diff --git "a/data/towardsai_com_pages.json" "b/data/towardsai_com_pages.json" new file mode 100644--- /dev/null +++ "b/data/towardsai_com_pages.json" @@ -0,0 +1,17748 @@ +{ + "schema_version": 2, + "source": "https://towardsai.com/pages-sitemap.xml", + "generated_at": "2026-08-09T11:24:43.395909+00:00", + "authority": "official_site", + "status_counts": { + "excluded": 3, + "included": 27 + }, + "pages": [ + { + "discovered_url": "https://towardsai.com/", + "url": "https://towardsai.com/", + "canonical_url": "https://towardsai.com/", + "host": "towardsai.com", + "path": "/", + "kind": "page", + "offer_id": null, + "entity_id": "page:towardsai.com/", + "title": "Towards AI · The AI deployment and education firm", + "review_title": "Homepage", + "reviewed_summary": "Towards AI deployment, education, and company overview.", + "meta_description": "Towards AI turns AI from a tool people experiment with into a capability teams rely on. We train AI engineers and workforces, and deploy custom AI systems inside enterprises, funds, and private equity portfolios.", + "headings": [ + "We train AI Engineers.", + "We deploy Enterprise AI.", + "Three ways to help you.", + "For Individuals", + "For Enterprise Workforce Trainings", + "For Enterprise Custom AI Deployments", + "Three groups we train .", + "From AI-curious to AI-capable, in your own time.", + "Where we create value .", + "AI deployments tied to EBITDA.", + "Three recent engagements .", + "One flywheel . Content to delivery.", + "Publication, Discord & Newsletter", + "Academy", + "Community", + "Enterprise Work", + "Building LLMs for Production. A practical reference inside real companies.", + "Technical depth and operator context." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Browse courses Talk to us\n\nWe train AI Engineers.\nBuilt by operators from / / / / / MILA Proprietary AI playbooks for Portfolio companies, Trading firms, and Asset Managers. Deep AI × Domain × Deployment expertise in one team. 500K+ AI practitioners taught since 2019. 200K+ Newsletter Subscribers (incl. Nvidia CEO Jensen Huang) . 100K+ Discord members. Join us .\n\nWe deploy Enterprise AI.\nBrowse our Learning Academy Browse our Learning Academy Talk to us about Custom Deployments Talk to us about Custom Deployments Trusted by Towards AI is the AI deployment and education firm . We turn AI from a tool people experiment with into a capability teams rely on. We believe AI initiatives fail when organisations lack the skills, education, and deployment capability to realise value, so we train your team while we build .\n\nThree ways to help you.\n\nFor Individuals\nAcademy: From developer to full-stack AI engineer Content: Newsletter, YouTube, LinkedIn Community: 100k+ members on Discord Explore community arrow_forward Browse courses arrow_forward\n\nFor Enterprise Workforce Trainings\nFoundations: AI literacy, policy, and governance Non-technical teams: 10x productivity with sub-agents Engineering teams: Become best-in-class AI engineers Leadership: AI strategy and value-creation playbooks Explore enablement arrow_forward Talk to us\n\nFor Enterprise Custom AI Deployments\nScope: Identify high-impact AI opportunities Build: Ship production-grade AI, not demos Sustain: Tools your team owns end-to-end Explore custom development arrow_forward Talk to us\n\nThree groups we train .\nOur enablement practice is built for distinct audiences. Hover any group. Individuals arrow_forward Enterprise / engineering teams arrow_forward Enterprise / non-technical staff arrow_forward For independent learners\n\nFrom AI-curious to AI-capable, in your own time.\nSelf-paced courses, our book that has sold 10,000+ copies and is used inside enterprise teams, our weekly newsletter, daily Discord, and a writer network of 4,000+ AI practitioners. Full-stack AI Engineer (1k+ courses sold!) Certifications recognised by hiring teams A community where you can actually ask questions and get answers\n\nWhere we create value .\nOur value-creation practice ships custom AI into operations. Hover any sector to see how. Private Equity arrow_forward Software Portco arrow_forward Industrials Portco arrow_forward Trading & Asset Management arrow_forward For PE-backed portfolios\n\nAI deployments tied to EBITDA.\nOur co-founders come from BCG and have worked inside leading PE firms. We partner with PE partners, operating partners, and portco CEOs to scope, build, and roll out AI use cases that show up on the next quarter’s P&L. We measure engineering proof, adoption, and financial impact separately. Diligence support and value-creation thesis pressure-testing Document automation, revenue-ops co-pilots, support deflection Cross-portfolio enablement so one playbook lifts every company Testimonials\n\nThree recent engagements .\n🇺🇸 Leading PE Fund HVAC services buy & build · USA Problem statement Workers logged material costs per job on paper. Backoffice manually re-typed from paper into Excel. Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails. What we built AI pipeline that takes PDFs of paper receipts Automatic extraction & calculation of material costs per Job Real-time consolidation & job-level reporting Audit-ready monthly reports Outcome Hours of finance team time saved per week. Audit-ready reports available instantly, not at month-end. 🇬🇧 Veloce IT IT services buy & build · UK Problem statement Multiple team members spent hours daily scanning fragmented tender sites across many geographies. Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most. What we built Automated scraping across 5+ tender platforms AI ranking by relevance for Veloce AI categorization of likely win-rate Daily report of newly-relevant opportunities Outcome Many hours of analyst time recovered per week. Pipeline focused on tenders Veloce is structurally most likely to win. 🇫🇷 Europol Law enforcement agency · France Problem statement Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines. What we built Multi-day intensive bootcamp on LLM pipelines Tailored to law-enforcement workflows Secure deployment for sensitive intelligence work Hands-on labs for context engineering Outcome Special agents trained and ready to deploy AI tools across operational intelligence workflows.\n\nOne flywheel . Content to delivery.\nMost AI firms are educators, consultants, or builders. Towards AI is all three. The flywheel keeps the company state-of-the-art and grounded in real outcomes. Stage 1 · Reach\n\nPublication, Discord & Newsletter\nWhere we attract practitioners check_circle 200k+ newsletter readers, including Jensen Huang check_circle 100k+ Discord members and 4,000+ writers Stage 2 · Convert\n\nAcademy\nWhere interest becomes skill check_circle 500,000+ learners taught since 2019 check_circle Self-paced courses, certifications, and cohort bootcamps Stage 3 · Compound\n\nCommunity\nWhere the practice strengthens check_circle Live Discord, writer network, and alumni from every cohort check_circle Patterns surface from real engineering pain, not theory Stage 4 · Deliver\n\nEnterprise Work\nWhere production teaches us check_circle Custom AI deployments and enterprise enablement check_circle Production lessons feed back into the curriculum Thought leadership\n\nBuilding LLMs for Production. A practical reference inside real companies.\nCo-written by our CTO Louis-François Bouchard and CEO Louie Peters, with contributions from 10+ Towards AI engineers and curation from MILA, Activeloop, and LlamaIndex. 465 pages on prompting, retrieval, fine-tuning, RAG, agents, and reliability. check_circle 10,000+ copies sold check_circle 4.4 stars on Amazon, 208 ratings check_circle Used as an internal reference inside enterprise AI teams “The most comprehensive textbook to date on building LLM applications. All essential topics in an AI engineer’s toolkit.” — Jerry Liu, Co-founder and CEO, LlamaIndex View on Amazon arrow_forward Leadership\n\nTechnical depth and operator context.\nLouie Peters CEO and Co-founder A leading voice on applied enterprise AI. Co-author of Building LLMs for Production, a 465-page reference used inside enterprise AI teams. Writes the weekly Towards AI newsletter to 200k+ readers. Former VP and Credit Research Lead Analyst at J.P. Morgan. Focus: practical AI transformation and adoption, cutting-edge AI system design and commercial strategy. Louis-François Bouchard CTO and Co-founder With over 200k+ followers across platforms, a recognized technical authority on AI engineering. Regular speaker at AI conferences like AI Engineer Europe, AI4, O’Reilly, the Uphill Conference. Co-author of the best-selling book Building LLMs for Production. Teaches AI engineering for O’Reilly and runs the What’s AI YouTube channel since 2020, one of the longest-running practitioner-facing AI channels online. Former Lead of AI at Designstripe and Technical AI Writing Consultant at EY. Former PhD student at MILA, one of the world’s leading AI research institutes. Denis Piffaretti Co-founder Co-founder at Empower, an acquisition-led UK building technical services platform and founding advisor at Veloce IT, an acquisition-led IT services platform. Former VP in Special Situations Trading at J.P. Morgan. Brings PE-backed operator and roll-up experience to the Towards AI Enterprise and PE-backed Value Creation motion. Start building with us Whatever you need from AI — learning, training, or building — you land with the same team of senior AI engineers and instructors. Talk to us arrow_forward Browse courses", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation" + }, + { + "text": "Mentorship", + "url": "https://towardsai.com/academy/mentorship/" + }, + { + "text": "Discord", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Browse courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Join us", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Browse our Learning Academy Browse our Learning Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Explore community arrow_forward", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Browse courses arrow_forward", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Explore enablement arrow_forward", + "url": "https://towardsai.com/enterpriseenablement" + }, + { + "text": "Explore custom development arrow_forward", + "url": "https://towardsai.com/valuecreation" + }, + { + "text": "View on Amazon arrow_forward", + "url": "https://www.amazon.com/Building-LLMs-Production-Enhancing-Capabilities/dp/B0D4FFPFW8" + }, + { + "text": "", + "url": "https://www.linkedin.com/in/louie-peters/" + }, + { + "text": "", + "url": "https://x.com/_LouiePeters" + }, + { + "text": "", + "url": "https://www.linkedin.com/in/whats-ai/" + }, + { + "text": "", + "url": "https://x.com/Whats_AI" + }, + { + "text": "", + "url": "https://www.linkedin.com/in/denis-p-72588a44/" + } + ], + "chunks": [ + { + "chunk_id": "tai-18cbf05fb6f9df482b46d2f0", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Browse courses Talk to us\n\nWe train AI Engineers.\nBuilt by operators from / / / / / MILA Proprietary AI playbooks for Portfolio companies, Trading firms, and Asset Managers. Deep AI × Domain × Deployment expertise in one team. 500K+ AI practitioners taught since 2019. 200K+ Newsletter Subscribers (incl. Nvidia CEO Jensen Huang) . 100K+ Discord members. Join us .\n\nWe deploy Enterprise AI.\nBrowse our Learning Academy Browse our Learning Academy Talk to us about Custom Deployments Talk to us about Custom Deployments Trusted by Towards AI is the AI deployment and education firm . We turn AI from a tool people experiment with into a capability teams rely on. We believe AI initiatives fail when organisations lack the skills, education, and deployment capability to realise value, so we train your team while we build .\n\nThree ways to help you.\n\nFor Individuals\nAcademy: From developer to full-stack AI engineer Content: Newsletter, YouTube, LinkedIn Community: 100k+ members on Discord Explore community arrow_forward Browse courses arrow_forward\n\nFor Enterprise Workforce Trainings\nFoundations: AI literacy, policy, and governance Non-technical teams: 10x productivity with sub-agents Engineering teams: Become best-in-class AI engineers Leadership: AI strategy and value-creation playbooks Explore enablement arrow_forward Talk to us", + "heading": "We deploy Enterprise AI.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-9836c3e6bfb5e57512d6b8d8", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-311c4c5e48e2ee3bef71d729", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-2d2b11c096850731a52e1647", + "text": "Community keyboard_arrow_down Mentorship Discord" + }, + { + "span_id": "span-05f0dbc71e92a917ff8b3a46", + "text": "Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-5afe751a68694657cb575e0e", + "text": "Browse courses Talk to us" + }, + { + "span_id": "span-09c43fb38e6503d26f731b27", + "text": "Built by operators from" + }, + { + "span_id": "span-f300b9cf9a8d6f3400926d24", + "text": "Proprietary AI playbooks for Portfolio companies," + }, + { + "span_id": "span-139e2bb3dd438b3febf9ada1", + "text": "Trading firms, and Asset Managers." + }, + { + "span_id": "span-f47b26aa7fe3c331cd4bf4d2", + "text": "Deep AI × Domain × Deployment expertise in one team." + }, + { + "span_id": "span-33db7382ca86c64a60724eae", + "text": "500K+ AI practitioners taught since 2019." + }, + { + "span_id": "span-350c54102a80fc3f4015e247", + "text": "200K+ Newsletter Subscribers" + }, + { + "span_id": "span-4b8d3543d452603195bcf2be", + "text": "Nvidia CEO Jensen Huang) ." + }, + { + "span_id": "span-c244f23acec687a1983cda53", + "text": "100K+ Discord members." + }, + { + "span_id": "span-a756cc5bf94a5427c7c5d43a", + "text": "Join us ." + }, + { + "span_id": "span-c243d352205f65345c549600", + "text": "Browse our" + }, + { + "span_id": "span-373a1ad3f106a1088f499e1f", + "text": "Learning Academy Browse our" + }, + { + "span_id": "span-0e408bdd2ec01e4be2ad6adf", + "text": "Learning Academy Talk to us about" + }, + { + "span_id": "span-90f157926a8cda147972f296", + "text": "Custom Deployments Talk to us about" + }, + { + "span_id": "span-a0780462fa2089c4fd62cdd8", + "text": "Custom Deployments" + }, + { + "span_id": "span-d1cbca5d5190969a804f4662", + "text": "Trusted by" + }, + { + "span_id": "span-ed210aa40881fb4763cd9b1a", + "text": "Towards AI is the AI deployment and education firm ." + }, + { + "span_id": "span-a7cd759231b58773aec6c943", + "text": "We turn AI from a tool people experiment with into a capability teams rely on." + }, + { + "span_id": "span-266503dae752cd79ce8ccdab", + "text": "We believe AI initiatives fail when organisations lack the skills, education, and deployment capability to realise value, so we train your team while we build ." + }, + { + "span_id": "span-2aac60bfd1eee5bd0634f745", + "text": "Three ways to help you." + }, + { + "span_id": "span-a74afd21de140ebbddb53675", + "text": "Academy: From developer to full-stack AI engineer" + }, + { + "span_id": "span-58ddb08f9274dcad1c1c5b34", + "text": "Content: Newsletter, YouTube, LinkedIn" + }, + { + "span_id": "span-7df818cc0a5fdb8a8b1b1e46", + "text": "Community: 100k+ members on Discord" + }, + { + "span_id": "span-9e4e4d306dbea12d72c7ca48", + "text": "Explore community arrow_forward Browse courses arrow_forward" + }, + { + "span_id": "span-69d3af0ea3561a5eb5306cf6", + "text": "Foundations: AI literacy, policy, and governance" + }, + { + "span_id": "span-ee7dc4f1c2f634a6e93a5da6", + "text": "Non-technical teams: 10x productivity with sub-agents" + }, + { + "span_id": "span-55d2aa1153496712b943ce10", + "text": "Engineering teams: Become best-in-class AI engineers" + }, + { + "span_id": "span-59ed2d282a6406279c05d4e7", + "text": "Leadership: AI strategy and value-creation playbooks" + }, + { + "span_id": "span-bb994866221fa08036c68c51", + "text": "Explore enablement arrow_forward Talk to us" + } + ] + }, + { + "chunk_id": "tai-f991c78dec95d804fd359d7d", + "text": "For Enterprise Custom AI Deployments\nScope: Identify high-impact AI opportunities Build: Ship production-grade AI, not demos Sustain: Tools your team owns end-to-end Explore custom development arrow_forward Talk to us\n\nThree groups we train .\nOur enablement practice is built for distinct audiences. Hover any group. Individuals arrow_forward Enterprise / engineering teams arrow_forward Enterprise / non-technical staff arrow_forward For independent learners\n\nFrom AI-curious to AI-capable, in your own time.\nSelf-paced courses, our book that has sold 10,000+ copies and is used inside enterprise teams, our weekly newsletter, daily Discord, and a writer network of 4,000+ AI practitioners. Full-stack AI Engineer (1k+ courses sold!) Certifications recognised by hiring teams A community where you can actually ask questions and get answers\n\nWhere we create value .\nOur value-creation practice ships custom AI into operations. Hover any sector to see how. Private Equity arrow_forward Software Portco arrow_forward Industrials Portco arrow_forward Trading & Asset Management arrow_forward For PE-backed portfolios\n\nAI deployments tied to EBITDA.\nOur co-founders come from BCG and have worked inside leading PE firms. We partner with PE partners, operating partners, and portco CEOs to scope, build, and roll out AI use cases that show up on the next quarter’s P&L. We measure engineering proof, adoption, and financial impact separately. Diligence support and value-creation thesis pressure-testing Document automation, revenue-ops co-pilots, support deflection Cross-portfolio enablement so one playbook lifts every company Testimonials", + "heading": "AI deployments tied to EBITDA.", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-422d833a21029cdc1d794c49", + "text": "Scope: Identify high-impact AI opportunities" + }, + { + "span_id": "span-93a4f94881b957de46a23346", + "text": "Build: Ship production-grade AI, not demos" + }, + { + "span_id": "span-44af0d20aaf5bf0cd079de07", + "text": "Sustain: Tools your team owns end-to-end" + }, + { + "span_id": "span-e44591d6c1155f4640a1fe3f", + "text": "Explore custom development arrow_forward Talk to us" + }, + { + "span_id": "span-3e165cc32203fe038d0d5e1f", + "text": "Our enablement practice is built for distinct audiences." + }, + { + "span_id": "span-175bf5cf985ff020e679fa4f", + "text": "Hover any group." + }, + { + "span_id": "span-77e787ccb98a96e078904915", + "text": "Individuals arrow_forward Enterprise / engineering teams arrow_forward Enterprise / non-technical staff arrow_forward" + }, + { + "span_id": "span-114856f7190b749a2f6026c7", + "text": "For independent learners" + }, + { + "span_id": "span-a0aebe9d7038f7a18a29a6de", + "text": "Self-paced courses, our book that has sold 10,000+ copies and is used inside enterprise teams, our weekly newsletter, daily Discord, and a writer network of 4,000+ AI practitioners." + }, + { + "span_id": "span-9c7dbf257593417bb5f15b9d", + "text": "Full-stack AI Engineer (1k+ courses sold!)" + }, + { + "span_id": "span-ddd34d4d7d16232da801beda", + "text": "Certifications recognised by hiring teams" + }, + { + "span_id": "span-ead1b1fe69d281f7ac50f71e", + "text": "A community where you can actually ask questions and get answers" + }, + { + "span_id": "span-f60902e46cc86a0729a05c6f", + "text": "Our value-creation practice ships custom AI into operations." + }, + { + "span_id": "span-b6f8018698b090caa6b49f95", + "text": "Hover any sector to see how." + }, + { + "span_id": "span-2c2550dc31bd2a071b78bd85", + "text": "Private Equity arrow_forward Software Portco arrow_forward Industrials Portco arrow_forward Trading & Asset Management arrow_forward" + }, + { + "span_id": "span-c8b03fecf83fb3373c48758d", + "text": "For PE-backed portfolios" + }, + { + "span_id": "span-31d68d8634c32c2081556105", + "text": "Our co-founders come from BCG and have worked inside leading PE firms." + }, + { + "span_id": "span-e6147ecfaf7d3f702fb965f3", + "text": "We partner with PE partners, operating partners, and portco CEOs to scope, build, and roll out AI use cases that show up on the next quarter’s P&L." + }, + { + "span_id": "span-1a0638a1989ca09e3efcedd4", + "text": "We measure engineering proof, adoption, and financial impact separately." + }, + { + "span_id": "span-e7cfecb9770acb9a15fc2b5b", + "text": "Diligence support and value-creation thesis pressure-testing" + }, + { + "span_id": "span-cc7f1d053f46e91fae3b6445", + "text": "Document automation, revenue-ops co-pilots, support deflection" + }, + { + "span_id": "span-1c8b851c73a439bd5ce5375c", + "text": "Cross-portfolio enablement so one playbook lifts every company" + } + ] + }, + { + "chunk_id": "tai-659ad53343baeb91a6280bfe", + "text": "Three recent engagements .\n🇺🇸 Leading PE Fund HVAC services buy & build · USA Problem statement Workers logged material costs per job on paper. Backoffice manually re-typed from paper into Excel. Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails. What we built AI pipeline that takes PDFs of paper receipts Automatic extraction & calculation of material costs per Job Real-time consolidation & job-level reporting Audit-ready monthly reports Outcome Hours of finance team time saved per week. Audit-ready reports available instantly, not at month-end. 🇬🇧 Veloce IT IT services buy & build · UK Problem statement Multiple team members spent hours daily scanning fragmented tender sites across many geographies. Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most. What we built Automated scraping across 5+ tender platforms AI ranking by relevance for Veloce AI categorization of likely win-rate Daily report of newly-relevant opportunities Outcome Many hours of analyst time recovered per week. Pipeline focused on tenders Veloce is structurally most likely to win. 🇫🇷 Europol Law enforcement agency · France Problem statement Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines. What we built Multi-day intensive bootcamp on LLM pipelines Tailored to law-enforcement workflows Secure deployment for sensitive intelligence work Hands-on labs for context engineering Outcome Special agents trained and ready to deploy AI tools across operational intelligence workflows.", + "heading": "Three recent engagements .", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-c2a6475472f922584b7e3084", + "text": "🇺🇸 Leading PE Fund" + }, + { + "span_id": "span-f8ba121572472fc1f787ac02", + "text": "HVAC services buy & build · USA" + }, + { + "span_id": "span-313b44ea5c3ee35ee334c51e", + "text": "Problem statement" + }, + { + "span_id": "span-bd8cd386776a9c82a4e95cdf", + "text": "Workers logged material costs per job on paper." + }, + { + "span_id": "span-4bdd42c7afef4b8b09bba352", + "text": "Backoffice manually re-typed from paper into Excel." + }, + { + "span_id": "span-6a5273f509720efc44e2cab4", + "text": "Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails." + }, + { + "span_id": "span-ba8db956ddd3022fecdbdf22", + "text": "What we built" + }, + { + "span_id": "span-bcbbac1f8633aed7b4a6d2c9", + "text": "AI pipeline that takes PDFs of paper receipts" + }, + { + "span_id": "span-5724e7403d078c5215a8cbd1", + "text": "Automatic extraction & calculation of material costs per Job" + }, + { + "span_id": "span-89a4abb5c35d0dc60d99d310", + "text": "Real-time consolidation & job-level reporting" + }, + { + "span_id": "span-2292708d400ff6aad4ecc485", + "text": "Audit-ready monthly reports" + }, + { + "span_id": "span-52aa00155329a95ba342b613", + "text": "Hours of finance team time saved per week." + }, + { + "span_id": "span-529fa24f6b55b8bd448adf09", + "text": "Audit-ready reports available instantly, not at month-end." + }, + { + "span_id": "span-6e83efdf25aeec858a54e575", + "text": "🇬🇧 Veloce IT" + }, + { + "span_id": "span-8ef8f80232b501ac2337e1a4", + "text": "IT services buy & build · UK" + }, + { + "span_id": "span-40da76d4ef3981f815b9a924", + "text": "Multiple team members spent hours daily scanning fragmented tender sites across many geographies." + }, + { + "span_id": "span-adca4a09ed0d2994a2958f3d", + "text": "Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most." + }, + { + "span_id": "span-2e515a3c2983741410ca9c4d", + "text": "Automated scraping across 5+ tender platforms" + }, + { + "span_id": "span-bc8ec1ea7c482236074d785f", + "text": "AI ranking by relevance for Veloce" + }, + { + "span_id": "span-4e2d508edaec6d3d40384b0f", + "text": "AI categorization of likely win-rate" + }, + { + "span_id": "span-882d12030584218fa5eaf822", + "text": "Daily report of newly-relevant opportunities" + }, + { + "span_id": "span-9d0d673813f97de9e6a5c624", + "text": "Many hours of analyst time recovered per week." + }, + { + "span_id": "span-69256d96c5025157a3e58457", + "text": "Pipeline focused on tenders Veloce is structurally most likely to win." + }, + { + "span_id": "span-5f11f25ad6f0fa200f6a2b42", + "text": "Law enforcement agency · France" + }, + { + "span_id": "span-59b748ce2cbbd128352d0b69", + "text": "Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines." + }, + { + "span_id": "span-5d7ebe107b1e34751110ba8c", + "text": "Multi-day intensive bootcamp on LLM pipelines" + }, + { + "span_id": "span-f6ec1923ebccf3cd48b86ce0", + "text": "Tailored to law-enforcement workflows" + }, + { + "span_id": "span-50a92833e5df8e3eb99f28ee", + "text": "Secure deployment for sensitive intelligence work" + }, + { + "span_id": "span-786fe0de2b60101bf9aa6f55", + "text": "Hands-on labs for context engineering" + }, + { + "span_id": "span-1506596270100934456e0e0e", + "text": "Special agents trained and ready to deploy AI tools across operational intelligence workflows." + } + ] + }, + { + "chunk_id": "tai-60bc5376dae56f01ef7d0b34", + "text": "One flywheel . Content to delivery.\nMost AI firms are educators, consultants, or builders. Towards AI is all three. The flywheel keeps the company state-of-the-art and grounded in real outcomes. Stage 1 · Reach\n\nPublication, Discord & Newsletter\nWhere we attract practitioners check_circle 200k+ newsletter readers, including Jensen Huang check_circle 100k+ Discord members and 4,000+ writers Stage 2 · Convert\n\nAcademy\nWhere interest becomes skill check_circle 500,000+ learners taught since 2019 check_circle Self-paced courses, certifications, and cohort bootcamps Stage 3 · Compound\n\nCommunity\nWhere the practice strengthens check_circle Live Discord, writer network, and alumni from every cohort check_circle Patterns surface from real engineering pain, not theory Stage 4 · Deliver\n\nEnterprise Work\nWhere production teaches us check_circle Custom AI deployments and enterprise enablement check_circle Production lessons feed back into the curriculum Thought leadership\n\nBuilding LLMs for Production. A practical reference inside real companies.\nCo-written by our CTO Louis-François Bouchard and CEO Louie Peters, with contributions from 10+ Towards AI engineers and curation from MILA, Activeloop, and LlamaIndex. 465 pages on prompting, retrieval, fine-tuning, RAG, agents, and reliability. check_circle 10,000+ copies sold check_circle 4.4 stars on Amazon, 208 ratings check_circle Used as an internal reference inside enterprise AI teams “The most comprehensive textbook to date on building LLM applications. All essential topics in an AI engineer’s toolkit.” — Jerry Liu, Co-founder and CEO, LlamaIndex View on Amazon arrow_forward Leadership", + "heading": "Building LLMs for Production. A practical reference inside real companies.", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-f52e53636bf58943cf0c97ee", + "text": "Most AI firms are educators, consultants, or builders." + }, + { + "span_id": "span-b09ef04ba2e9058d732e4bce", + "text": "Towards AI is all three." + }, + { + "span_id": "span-0d3f7f5b29f3bacecbf6b421", + "text": "The flywheel keeps the company state-of-the-art and grounded in real outcomes." + }, + { + "span_id": "span-313c1043f9af7fb19ce981ff", + "text": "Stage 1 · Reach" + }, + { + "span_id": "span-70ff60ec9026af489552b03b", + "text": "Where we attract practitioners" + }, + { + "span_id": "span-c6d4c8233c9f8576cf9688be", + "text": "check_circle 200k+ newsletter readers, including Jensen Huang" + }, + { + "span_id": "span-d8ea8ae8ea030d546a995b7e", + "text": "check_circle 100k+ Discord members and 4,000+ writers" + }, + { + "span_id": "span-ca1f7d1fe3bba02ff6751e01", + "text": "Stage 2 · Convert" + }, + { + "span_id": "span-ec83297b45e322622eb49604", + "text": "Where interest becomes skill" + }, + { + "span_id": "span-4ac57320f9eda9dbaa563228", + "text": "check_circle 500,000+ learners taught since 2019" + }, + { + "span_id": "span-a9c99dc82db4a6e449e8be4b", + "text": "check_circle Self-paced courses, certifications, and cohort bootcamps" + }, + { + "span_id": "span-4dfd266ab6a4a158bc31a7ae", + "text": "Stage 3 · Compound" + }, + { + "span_id": "span-b337973c4ceeac48a2b41804", + "text": "Where the practice strengthens" + }, + { + "span_id": "span-893b51a7b1aa8f823caf3621", + "text": "check_circle Live Discord, writer network, and alumni from every cohort" + }, + { + "span_id": "span-fd818bcd1648c25016174ecd", + "text": "check_circle Patterns surface from real engineering pain, not theory" + }, + { + "span_id": "span-fba4211788a4261bef6df112", + "text": "Stage 4 · Deliver" + }, + { + "span_id": "span-e3199ffee73e8cc8a541bc03", + "text": "Where production teaches us" + }, + { + "span_id": "span-310fbc026d6cb71924623bfa", + "text": "check_circle Custom AI deployments and enterprise enablement" + }, + { + "span_id": "span-18f4b39fb420103fddf1e64c", + "text": "check_circle Production lessons feed back into the curriculum" + }, + { + "span_id": "span-419f6c932b1f1a67dcf407d4", + "text": "Thought leadership" + }, + { + "span_id": "span-9e7a7774814565a9f9d61db3", + "text": "Co-written by our CTO Louis-François Bouchard and CEO Louie Peters, with contributions from 10+ Towards AI engineers and curation from MILA, Activeloop, and LlamaIndex." + }, + { + "span_id": "span-b3e98c4579d412b60144ea7e", + "text": "465 pages on prompting, retrieval, fine-tuning, RAG, agents, and reliability." + }, + { + "span_id": "span-8ca2dac4b41e9f7fdb2d2879", + "text": "check_circle 10,000+ copies sold" + }, + { + "span_id": "span-a70b3107f500f1f092e05152", + "text": "check_circle 4.4 stars on Amazon, 208 ratings" + }, + { + "span_id": "span-7993641cb667d5dd8bcc80bb", + "text": "check_circle Used as an internal reference inside enterprise AI teams" + }, + { + "span_id": "span-e7d559e4f8898c2837094a44", + "text": "“The most comprehensive textbook to date on building LLM applications." + }, + { + "span_id": "span-b6b07929bb42f621ebd8ee3e", + "text": "All essential topics in an AI engineer’s toolkit.” — Jerry Liu, Co-founder and CEO, LlamaIndex" + }, + { + "span_id": "span-d444f6253141459444448001", + "text": "View on Amazon arrow_forward" + } + ] + }, + { + "chunk_id": "tai-ffb203de5b2c81c82086aa8a", + "text": "Technical depth and operator context.\nLouie Peters CEO and Co-founder A leading voice on applied enterprise AI. Co-author of Building LLMs for Production, a 465-page reference used inside enterprise AI teams. Writes the weekly Towards AI newsletter to 200k+ readers. Former VP and Credit Research Lead Analyst at J.P. Morgan. Focus: practical AI transformation and adoption, cutting-edge AI system design and commercial strategy. Louis-François Bouchard CTO and Co-founder With over 200k+ followers across platforms, a recognized technical authority on AI engineering. Regular speaker at AI conferences like AI Engineer Europe, AI4, O’Reilly, the Uphill Conference. Co-author of the best-selling book Building LLMs for Production. Teaches AI engineering for O’Reilly and runs the What’s AI YouTube channel since 2020, one of the longest-running practitioner-facing AI channels online. Former Lead of AI at Designstripe and Technical AI Writing Consultant at EY. Former PhD student at MILA, one of the world’s leading AI research institutes. Denis Piffaretti Co-founder Co-founder at Empower, an acquisition-led UK building technical services platform and founding advisor at Veloce IT, an acquisition-led IT services platform. Former VP in Special Situations Trading at J.P. Morgan. Brings PE-backed operator and roll-up experience to the Towards AI Enterprise and PE-backed Value Creation motion. Start building with us Whatever you need from AI — learning, training, or building — you land with the same team of senior AI engineers and instructors. Talk to us arrow_forward Browse courses", + "heading": "Technical depth and operator context.", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-18dd8c82150cb3341ba643ae", + "text": "Louie Peters" + }, + { + "span_id": "span-5ec2c3a5f026d583e34c065c", + "text": "CEO and Co-founder" + }, + { + "span_id": "span-5bc696b14d89f74a675da612", + "text": "A leading voice on applied enterprise AI." + }, + { + "span_id": "span-db443015ea0b0ac9576f75ce", + "text": "Co-author of Building LLMs for Production, a 465-page reference used inside enterprise AI teams." + }, + { + "span_id": "span-f79e3794f6dee0b1ca4821fd", + "text": "Writes the weekly Towards AI newsletter to 200k+ readers." + }, + { + "span_id": "span-59496b064b2cc7b1651a0799", + "text": "Former VP and Credit Research Lead Analyst at J.P." + }, + { + "span_id": "span-012e78d4c96df93fbe50f831", + "text": "Focus: practical AI transformation and adoption, cutting-edge AI system design and commercial strategy." + }, + { + "span_id": "span-2dbf47e304efe47b23457730", + "text": "Louis-François Bouchard" + }, + { + "span_id": "span-4418501a71a0b5bc3e185303", + "text": "CTO and Co-founder" + }, + { + "span_id": "span-72f33e68d64c7b0db461735e", + "text": "With over 200k+ followers across platforms, a recognized technical authority on AI engineering." + }, + { + "span_id": "span-7e7fda23acaa0baf6993dd6b", + "text": "Regular speaker at AI conferences like AI Engineer Europe, AI4, O’Reilly, the Uphill Conference." + }, + { + "span_id": "span-beb94452ea24ceca3f374b7a", + "text": "Co-author of the best-selling book Building LLMs for Production." + }, + { + "span_id": "span-d3b877b75ed37879f0d70d8e", + "text": "Teaches AI engineering for O’Reilly and runs the What’s AI YouTube channel since 2020, one of the longest-running practitioner-facing AI channels online." + }, + { + "span_id": "span-eda1ac8f97d70bf96c5cc330", + "text": "Former Lead of AI at Designstripe and Technical AI Writing Consultant at EY." + }, + { + "span_id": "span-7302816fa080c06411d967d7", + "text": "Former PhD student at MILA, one of the world’s leading AI research institutes." + }, + { + "span_id": "span-40ab694945eebdaad3bb8858", + "text": "Denis Piffaretti" + }, + { + "span_id": "span-a4c6d686b5ec936ff4947eee", + "text": "Co-founder" + }, + { + "span_id": "span-d8642af371f0fc7a4d893634", + "text": "Co-founder at Empower, an acquisition-led UK building technical services platform and founding advisor at Veloce IT, an acquisition-led IT services platform." + }, + { + "span_id": "span-291b667ed02d7b35d41d1fbc", + "text": "Former VP in Special Situations Trading at J.P." + }, + { + "span_id": "span-e4e82d7c7163ac5d5538948b", + "text": "Brings PE-backed operator and roll-up experience to the Towards AI Enterprise and PE-backed Value Creation motion." + }, + { + "span_id": "span-a49a8cdac40879809da9a9b0", + "text": "Start building with us" + }, + { + "span_id": "span-601e1b7ee1d0b3dcad945602", + "text": "Whatever you need from AI — learning, training, or building — you land with the same team of senior AI engineers and instructors." + }, + { + "span_id": "span-e3876d354055b998fa3ed547", + "text": "Talk to us arrow_forward Browse courses" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "8a6d2a1fdd20254a1792041a6cb8ac301b4a12dd4a4089238317cc5539b6522e", + "content_hash": "8a6d2a1fdd20254a1792041a6cb8ac301b4a12dd4a4089238317cc5539b6522e", + "authority": "official_site", + "status": "excluded", + "retrieval_eligible": false, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/" + ], + "excluded_reason": "homepage offer summaries can conflict with canonical detail pages", + "evidence_hash": "3dad1c25ae60b8e64f3da2a4437cd21f69e55952ddc33da6e1fce210c03e348d" + }, + { + "discovered_url": "https://towardsai.com/academy/", + "url": "https://towardsai.com/academy/", + "canonical_url": "https://towardsai.com/academy/", + "host": "towardsai.com", + "path": "/academy", + "kind": "collection", + "offer_id": null, + "entity_id": "page:towardsai.com/academy", + "title": "Towards AI Academy · Build a Career in AI Engineering", + "review_title": "Academy hub", + "reviewed_summary": "Central comparison page for Towards AI learning paths.", + "meta_description": "Self-paced, industry-focused AI engineering courses for developers, professionals, and companies looking to become AI-first. 10,000+ engineers enrolled.", + "headings": [ + "Build a career in AI engineering", + "Industry leaders on our resources", + "The AI industry has a people problem", + "Learn to engineer what goes on top", + "Build judgment that outlasts any framework", + "Get trained in exactly what they're hiring for", + "You're in the right place if", + "Hear from our alumni", + "Why developers choose Towards AI", + "Created by 15+ senior AI engineers", + "Training the engineers the AI economy needs", + "Community, lifetime access, ongoing support", + "Your AI engineering pathway", + "Full Stack AI Engineering", + "Agent Engineering", + "Fully Video LLM Primer", + "Beginner Python for AI Engineering", + "Master AI for Productivity", + "Ready to become an AI engineer?", + "Frequently asked questions" + ], + "text": "Towards AI Academy\n\nBuild a career in AI engineering\nThrough project-based courses built by engineers who do this for a living. Become an AI Engineer Explore the pathway 10,000+ engineers enrolled across our courses 100k+ members in our Discord community 15+ senior AI engineers behind the curriculum 01 Trusted voices\n\nIndustry leaders on our resources\n02 Why this matters\n\nThe AI industry has a people problem\nWe're training the engineers it needs. Foundation models are just the starting point\n\nLearn to engineer what goes on top\nMaster prompting, RAG, fine-tuning, tool use, and agents, and build the scalable systems foundation models alone can't deliver. The skill gap is widening every quarter\n\nBuild judgment that outlasts any framework\nEvery program is built around production-grade projects. You learn to handle failures, optimize at scale, and ship reliably. Enterprises want AI. Engineers are the bottleneck.\n\nGet trained in exactly what they're hiring for\nA deployed product and a certified portfolio are built into every course, matched to the roles enterprises are hiring for. Get the AI training the industry needs 03 Is this for you?\n\nYou're in the right place if\n01 You want to ship a production-grade LLM product, and a notebook demo won't cut it. 02 You want to be the AI engineer your company can't afford to lose. 03 You want a project, a portfolio, and certification you can stand behind. 04 You want to master LLMs, RAG, fine-tuning, and agents, end-to-end. 05 You're building toward a role, a product, or an AI initiative worth leading. 06 You're willing to put in the work that most developers won't. 04 Student outcomes\n\nHear from our alumni\n05 Our difference\n\nWhy developers choose Towards AI\nTraining built for the industry\n\nCreated by 15+ senior AI engineers\nBuilt from what we implement for enterprise clients. You graduate with a deployed product, a certified portfolio, and the production experience to back it up. Designed to make developers AI-first\n\nTraining the engineers the AI economy needs\nSince 2019, our learners have moved into six-figure roles and led AI-first transformations, with a certified portfolio that proves what they can build. Created for working professionals\n\nCommunity, lifetime access, ongoing support\nSelf-paced with lifetime access, live kick-off calls, weekly updates, and a 100k+ Discord community, so you can learn on your schedule without falling behind. 06 The courses\n\nYour AI engineering pathway\nWhether you're new to coding, new to AI, or among the 10 million building with LLM APIs, this is your path to production-grade AI systems that actually work. 1 / 5 Flagship certification\n\nFull Stack AI Engineering\nBest for Python developers ready to master the full AI stack Taught by experts who do this for a living, our flagship, 60-hour, code-heavy program covers every layer of modern AI engineering. With a central 60-lesson context engineering project, it spans prompting, advanced RAG, fine-tuning, agents, and full system design. Weekly updates make this the only course you'll ever need. Learn more 2 / 5 Production agents\n\nAgent Engineering\nBest for Python developers ready to build fully autonomous agents A production-focused program covering the complete agent engineering stack. Through two capstone projects, a Research Agent and a Writing Workflow, you'll master agentic design, reasoning loops, tool orchestration, evaluation, and deployment. Created in partnership with Paul Iusztin. Learn more 3 / 5 Crash course\n\nFully Video LLM Primer\nBest for busy software professionals needing an LLM crash course Learn how to understand, build, evaluate, automate, and maintain robust LLM systems without getting bogged down in theory. In five focused video sessions, you'll learn when to use prompting, RAG, fine-tuning, or agents to build production-grade LLM applications. Learn more 4 / 5 Foundations\n\nBeginner Python for AI Engineering\nBest for non-coders and Python beginners starting with AI engineering Learn the Python skills for building and integrating AI applications, focused on what matters for LLMs. Ideal for those who need Python before our AI Engineering course, so you can create, test, and deploy faster. Learn more 5 / 5 No code required\n\nMaster AI for Productivity\nBest for professionals using AI for high-value tasks without coding Discover how to integrate AI safely and effectively into your daily work to save time, improve output quality, and open new opportunities. Work through practical, real-world scenarios across research, communication, analysis, and automation. Learn more Organizations we have worked with Start today\n\nReady to become an AI engineer?\nSelf-paced, project-based, and updated as fast as the field moves. Join 10,000+ engineers who made the leap. Begin your transition Try the free email course 07 Questions\n\nFrequently asked questions\nDo the courses come with a certification? + Yes, all courses come with a certificate of completion from Towards AI. What are the prerequisites for the resources? + Our Full Stack and Agentic Engineering courses require Python understanding. If you are a complete beginner, you can start with our Python course. What makes Towards AI different? + Every course is built directly from what our engineers implement for enterprise clients, so the curriculum reflects real production constraints. You learn by building production-grade systems end-to-end, graduate with a deployed product and certified portfolio, and join a curriculum that updates every month as the field evolves. How can I apply what I learn in my career? + The hands-on projects you complete during the course will help you build a portfolio that showcases your abilities to potential employers. Many of our learners have moved into high-paying AI roles or launched their own products. Can I get a discount? + Yes, there are three ways to receive a discount on our courses. If you're a student, we offer a 50% discount on all courses. If you've previously purchased our book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more people. Email Louis-François at louis@towardsai.net for a follow-up. Does Towards AI offer an Affiliate or Referral Program? + Yes. Towards AI offers an Affiliate Program and Referral Program, enabling you to earn up to 30% per sale and free access to our courses. See the Affiliate page for tiers and payouts. Do I get additional support throughout my learning journey? + Yes, you will get access to dedicated course channels in our Discord community, exclusive Slack access, and an AI Tutor chatbot. When were the courses last updated? + All our courses are updated every two weeks with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field instead of freezing at launch.", + "links": [ + { + "text": "Become an AI Engineer", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/academy/agent-engineering/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/academy/python-for-ai-engineering/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/academy/ai-for-work/" + }, + { + "text": "Try the free email course", + "url": "https://email-course.towardsai.net/" + }, + { + "text": "Affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + } + ], + "chunks": [ + { + "chunk_id": "tai-8a682c0ccf315793aa8390ae", + "text": "Towards AI Academy\n\nBuild a career in AI engineering\nThrough project-based courses built by engineers who do this for a living. Become an AI Engineer Explore the pathway 10,000+ engineers enrolled across our courses 100k+ members in our Discord community 15+ senior AI engineers behind the curriculum 01 Trusted voices\n\nIndustry leaders on our resources\n02 Why this matters\n\nThe AI industry has a people problem\nWe're training the engineers it needs. Foundation models are just the starting point\n\nLearn to engineer what goes on top\nMaster prompting, RAG, fine-tuning, tool use, and agents, and build the scalable systems foundation models alone can't deliver. The skill gap is widening every quarter\n\nBuild judgment that outlasts any framework\nEvery program is built around production-grade projects. You learn to handle failures, optimize at scale, and ship reliably. Enterprises want AI. Engineers are the bottleneck.\n\nGet trained in exactly what they're hiring for\nA deployed product and a certified portfolio are built into every course, matched to the roles enterprises are hiring for. Get the AI training the industry needs 03 Is this for you?\n\nYou're in the right place if\n01 You want to ship a production-grade LLM product, and a notebook demo won't cut it. 02 You want to be the AI engineer your company can't afford to lose. 03 You want a project, a portfolio, and certification you can stand behind. 04 You want to master LLMs, RAG, fine-tuning, and agents, end-to-end. 05 You're building toward a role, a product, or an AI initiative worth leading. 06 You're willing to put in the work that most developers won't. 04 Student outcomes", + "heading": "You're in the right place if", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-361318fcfa17fa3988776861", + "text": "Towards AI Academy" + }, + { + "span_id": "span-83e0066110d5443a523d485c", + "text": "Through project-based courses built by engineers who do this for a living." + }, + { + "span_id": "span-a44050b46d5bfc9181747477", + "text": "Become an AI Engineer Explore the pathway" + }, + { + "span_id": "span-f74944bfff2e3a5384d12190", + "text": "10,000+ engineers enrolled across our courses" + }, + { + "span_id": "span-b5a5456fca1aaddc0b3d55b0", + "text": "100k+ members in our Discord community" + }, + { + "span_id": "span-68292b8a4f05ad8fcde1236e", + "text": "15+ senior AI engineers behind the curriculum" + }, + { + "span_id": "span-fc5734dab4fc3a0ed58f7996", + "text": "01 Trusted voices" + }, + { + "span_id": "span-b15a0bbc3392ffafd54aab61", + "text": "02 Why this matters" + }, + { + "span_id": "span-1f1d939958068a4fc53e0b5c", + "text": "We're training the engineers it needs." + }, + { + "span_id": "span-1810082b9cfd3eb07537bac5", + "text": "Foundation models are just the starting point" + }, + { + "span_id": "span-ab941cebeb9d1cc323b53720", + "text": "Master prompting, RAG, fine-tuning, tool use, and agents, and build the scalable systems foundation models alone can't deliver." + }, + { + "span_id": "span-50150e122e81f690b95e698e", + "text": "The skill gap is widening every quarter" + }, + { + "span_id": "span-1a3af41542bec1644ba15715", + "text": "Every program is built around production-grade projects." + }, + { + "span_id": "span-0f581ec7a4a18059bc505845", + "text": "You learn to handle failures, optimize at scale, and ship reliably." + }, + { + "span_id": "span-0cda3e70ec90d05085c3cb17", + "text": "Enterprises want AI." + }, + { + "span_id": "span-02fcdca32834e0cf05e99e01", + "text": "Engineers are the bottleneck." + }, + { + "span_id": "span-8221f79affc8441faa9cc1b6", + "text": "A deployed product and a certified portfolio are built into every course, matched to the roles enterprises are hiring for." + }, + { + "span_id": "span-7a105a873f2eb8cfa9d28425", + "text": "Get the AI training the industry needs" + }, + { + "span_id": "span-b5218104eac1ed84a859d7f6", + "text": "03 Is this for you?" + }, + { + "span_id": "span-9d630dd9ab4f8a12b1379946", + "text": "01 You want to ship a production-grade LLM product, and a notebook demo won't cut it." + }, + { + "span_id": "span-54daa7f0bcee6574a33194bd", + "text": "02 You want to be the AI engineer your company can't afford to lose." + }, + { + "span_id": "span-ffbc087aefd5ecf71ce2a598", + "text": "03 You want a project, a portfolio, and certification you can stand behind." + }, + { + "span_id": "span-cb339a4e0737e21c6e9ecd04", + "text": "04 You want to master LLMs, RAG, fine-tuning, and agents, end-to-end." + }, + { + "span_id": "span-1a195282fc6e8575737554ae", + "text": "05 You're building toward a role, a product, or an AI initiative worth leading." + }, + { + "span_id": "span-2635b27e7d7e117904ebac3f", + "text": "06 You're willing to put in the work that most developers won't." + }, + { + "span_id": "span-521053252a6aeb9aff3d2d44", + "text": "04 Student outcomes" + } + ] + }, + { + "chunk_id": "tai-8649d343ad87ef317acbbe13", + "text": "Hear from our alumni\n05 Our difference\n\nWhy developers choose Towards AI\nTraining built for the industry\n\nCreated by 15+ senior AI engineers\nBuilt from what we implement for enterprise clients. You graduate with a deployed product, a certified portfolio, and the production experience to back it up. Designed to make developers AI-first\n\nTraining the engineers the AI economy needs\nSince 2019, our learners have moved into six-figure roles and led AI-first transformations, with a certified portfolio that proves what they can build. Created for working professionals\n\nCommunity, lifetime access, ongoing support\nSelf-paced with lifetime access, live kick-off calls, weekly updates, and a 100k+ Discord community, so you can learn on your schedule without falling behind. 06 The courses\n\nYour AI engineering pathway\nWhether you're new to coding, new to AI, or among the 10 million building with LLM APIs, this is your path to production-grade AI systems that actually work. 1 / 5 Flagship certification\n\nFull Stack AI Engineering\nBest for Python developers ready to master the full AI stack Taught by experts who do this for a living, our flagship, 60-hour, code-heavy program covers every layer of modern AI engineering. With a central 60-lesson context engineering project, it spans prompting, advanced RAG, fine-tuning, agents, and full system design. Weekly updates make this the only course you'll ever need. Learn more 2 / 5 Production agents", + "heading": "Full Stack AI Engineering", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-ae5370ac1eb0c642a57d9efb", + "text": "05 Our difference" + }, + { + "span_id": "span-5ce55808aa62a30c7ca0f96d", + "text": "Training built for the industry" + }, + { + "span_id": "span-799d61cff3387e0cb8cac1d4", + "text": "Created by 15+ senior AI engineers" + }, + { + "span_id": "span-d16d60c3ebef2da46c5f402d", + "text": "Built from what we implement for enterprise clients." + }, + { + "span_id": "span-8f0f9818add5eac540ad086d", + "text": "You graduate with a deployed product, a certified portfolio, and the production experience to back it up." + }, + { + "span_id": "span-dcd0030d206d83a564621928", + "text": "Designed to make developers AI-first" + }, + { + "span_id": "span-cb1bf76546ddc1df70eb0e0e", + "text": "Since 2019, our learners have moved into six-figure roles and led AI-first transformations, with a certified portfolio that proves what they can build." + }, + { + "span_id": "span-3ebb79eb3a10d253f7786004", + "text": "Created for working professionals" + }, + { + "span_id": "span-49f0bf48a9a8a9369898310f", + "text": "Self-paced with lifetime access, live kick-off calls, weekly updates, and a 100k+ Discord community, so you can learn on your schedule without falling behind." + }, + { + "span_id": "span-8acf54cedae094defa0939dd", + "text": "06 The courses" + }, + { + "span_id": "span-97c11c0f5e8dd8c46e5a1f54", + "text": "Whether you're new to coding, new to AI, or among the 10 million building with LLM APIs, this is your path to production-grade AI systems that actually work." + }, + { + "span_id": "span-d6d1f9693d0aced05c5539c9", + "text": "1 / 5 Flagship certification" + }, + { + "span_id": "span-52af4a7cbdb75ab2d324d3fe", + "text": "Best for Python developers ready to master the full AI stack Taught by experts who do this for a living, our flagship, 60-hour, code-heavy program covers every layer of modern AI engineering." + }, + { + "span_id": "span-8471196103daa8486d914210", + "text": "With a central 60-lesson context engineering project, it spans prompting, advanced RAG, fine-tuning, agents, and full system design." + }, + { + "span_id": "span-b2d85197125d3935e762c032", + "text": "Weekly updates make this the only course you'll ever need." + }, + { + "span_id": "span-4a00e57e94caa4a497f893cb", + "text": "Learn more" + }, + { + "span_id": "span-883d9102be797850404c3579", + "text": "2 / 5 Production agents" + } + ] + }, + { + "chunk_id": "tai-4b3d3fea7e4531c1f17ae2e1", + "text": "Agent Engineering\nBest for Python developers ready to build fully autonomous agents A production-focused program covering the complete agent engineering stack. Through two capstone projects, a Research Agent and a Writing Workflow, you'll master agentic design, reasoning loops, tool orchestration, evaluation, and deployment. Created in partnership with Paul Iusztin. Learn more 3 / 5 Crash course\n\nFully Video LLM Primer\nBest for busy software professionals needing an LLM crash course Learn how to understand, build, evaluate, automate, and maintain robust LLM systems without getting bogged down in theory. In five focused video sessions, you'll learn when to use prompting, RAG, fine-tuning, or agents to build production-grade LLM applications. Learn more 4 / 5 Foundations\n\nBeginner Python for AI Engineering\nBest for non-coders and Python beginners starting with AI engineering Learn the Python skills for building and integrating AI applications, focused on what matters for LLMs. Ideal for those who need Python before our AI Engineering course, so you can create, test, and deploy faster. Learn more 5 / 5 No code required\n\nMaster AI for Productivity\nBest for professionals using AI for high-value tasks without coding Discover how to integrate AI safely and effectively into your daily work to save time, improve output quality, and open new opportunities. Work through practical, real-world scenarios across research, communication, analysis, and automation. Learn more Organizations we have worked with Start today\n\nReady to become an AI engineer?\nSelf-paced, project-based, and updated as fast as the field moves. Join 10,000+ engineers who made the leap. Begin your transition Try the free email course 07 Questions", + "heading": "Agent Engineering", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-32e68d0cc35d750917ed072c", + "text": "Best for Python developers ready to build fully autonomous agents A production-focused program covering the complete agent engineering stack." + }, + { + "span_id": "span-cc0c89d2ab0f115f99292fbc", + "text": "Through two capstone projects, a Research Agent and a Writing Workflow, you'll master agentic design, reasoning loops, tool orchestration, evaluation, and deployment." + }, + { + "span_id": "span-5fb1149e8de3dbf0956c9dd7", + "text": "Created in partnership with Paul Iusztin." + }, + { + "span_id": "span-78bbd8d9cd732e679a814666", + "text": "Learn more" + }, + { + "span_id": "span-7f74914bb37ea734f8893626", + "text": "3 / 5 Crash course" + }, + { + "span_id": "span-8a0275b1e97ab5261e8375ac", + "text": "Best for busy software professionals needing an LLM crash course Learn how to understand, build, evaluate, automate, and maintain robust LLM systems without getting bogged down in theory." + }, + { + "span_id": "span-8f2d8b1f4cf0481ed638a347", + "text": "In five focused video sessions, you'll learn when to use prompting, RAG, fine-tuning, or agents to build production-grade LLM applications." + }, + { + "span_id": "span-998f30cbc2fbbdb8f333409c", + "text": "4 / 5 Foundations" + }, + { + "span_id": "span-c537628db44d51d15bbf4073", + "text": "Best for non-coders and Python beginners starting with AI engineering Learn the Python skills for building and integrating AI applications, focused on what matters for LLMs." + }, + { + "span_id": "span-14fd4c1d54a88e85bbe87e9a", + "text": "Ideal for those who need Python before our AI Engineering course, so you can create, test, and deploy faster." + }, + { + "span_id": "span-4af8c2a23182c550b20f549d", + "text": "5 / 5 No code required" + }, + { + "span_id": "span-764d9bc0ff6173f2ccf0248b", + "text": "Best for professionals using AI for high-value tasks without coding Discover how to integrate AI safely and effectively into your daily work to save time, improve output quality, and open new opportunities." + }, + { + "span_id": "span-67065d190fca1fdda51cd093", + "text": "Work through practical, real-world scenarios across research, communication, analysis, and automation." + }, + { + "span_id": "span-256a83be67377cf39eb72810", + "text": "Organizations we have worked with" + }, + { + "span_id": "span-e899b57cabde6a1e152bee11", + "text": "Start today" + }, + { + "span_id": "span-29d44e8d5a18d6b715e8464e", + "text": "Self-paced, project-based, and updated as fast as the field moves." + }, + { + "span_id": "span-f6cc2c8fd12809e3f121bf12", + "text": "Join 10,000+ engineers who made the leap." + }, + { + "span_id": "span-e3e18f012178c1bf6bb4b0fe", + "text": "Begin your transition Try the free email course" + }, + { + "span_id": "span-3c999f3174b5547bead4f078", + "text": "07 Questions" + } + ] + }, + { + "chunk_id": "tai-6dfeaa6c2f29cef92ccbf589", + "text": "Frequently asked questions\nDo the courses come with a certification? + Yes, all courses come with a certificate of completion from Towards AI. What are the prerequisites for the resources? + Our Full Stack and Agentic Engineering courses require Python understanding. If you are a complete beginner, you can start with our Python course. What makes Towards AI different? + Every course is built directly from what our engineers implement for enterprise clients, so the curriculum reflects real production constraints. You learn by building production-grade systems end-to-end, graduate with a deployed product and certified portfolio, and join a curriculum that updates every month as the field evolves. How can I apply what I learn in my career? + The hands-on projects you complete during the course will help you build a portfolio that showcases your abilities to potential employers. Many of our learners have moved into high-paying AI roles or launched their own products. Can I get a discount? + Yes, there are three ways to receive a discount on our courses. If you're a student, we offer a 50% discount on all courses. If you've previously purchased our book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more people. Email Louis-François at louis@towardsai.net for a follow-up. Does Towards AI offer an Affiliate or Referral Program? + Yes. Towards AI offers an Affiliate Program and Referral Program, enabling you to earn up to 30% per sale and free access to our courses. See the Affiliate page for tiers and payouts. Do I get additional support throughout my learning journey? + Yes, you will get access to dedicated course channels in our Discord community, exclusive Slack access, and an AI Tutor chatbot. When were the", + "heading": "Frequently asked questions", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-f3aeec483240a42167f6c396", + "text": "Yes, all courses come with a certificate of completion from Towards AI." + }, + { + "span_id": "span-cb405554389f650ac240aea5", + "text": "Our Full Stack and Agentic Engineering courses require Python understanding." + }, + { + "span_id": "span-576e5b78ecf1d394c7de292d", + "text": "If you are a complete beginner, you can start with our Python course." + }, + { + "span_id": "span-328ae53747e61e3027a9addd", + "text": "The hands-on projects you complete during the course will help you build a portfolio that showcases your abilities to potential employers." + }, + { + "span_id": "span-d4cc1cd08513b728e8de7b82", + "text": "Many of our learners have moved into high-paying AI roles or launched their own products." + }, + { + "span_id": "span-1ab268b8a8fa9ba2e737b865", + "text": "Towards AI offers an Affiliate Program and Referral Program, enabling you to earn up to 30% per sale and free access to our courses." + }, + { + "span_id": "span-92bae40be2fa723c54dae459", + "text": "See the Affiliate page for tiers and payouts." + }, + { + "span_id": "span-44435499c1cbe211249e78a7", + "text": "Yes, you will get access to dedicated course channels in our Discord community, exclusive Slack access, and an AI Tutor chatbot." + } + ] + }, + { + "chunk_id": "tai-7d804ee62b82ae7b4d8335d4", + "text": "Frequently asked questions\ncourses last updated? + All our courses are updated every two weeks with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field instead of freezing at launch.", + "heading": "Frequently asked questions", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-4ff3aa7928536694c75845fc", + "text": "All our courses are updated every two weeks with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field instead of freezing at launch." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "bcdd5b29fb1a518395932292655a4aada1f1ed570b935220ee29a16cd7690cad", + "content_hash": "bcdd5b29fb1a518395932292655a4aada1f1ed570b935220ee29a16cd7690cad", + "authority": "official_site", + "status": "excluded", + "retrieval_eligible": false, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/" + ], + "excluded_reason": "catalog summary can conflict with canonical offer detail pages", + "evidence_hash": "bed26bbb6c331bf0188c3939ae7f1f603c3c646ae2afe5685c1452a1780e163c" + }, + { + "discovered_url": "https://towardsai.com/academy/full-stack-ai-engineering/", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/", + "canonical_url": "https://towardsai.com/academy/full-stack-ai-engineering/", + "host": "towardsai.com", + "path": "/academy/full-stack-ai-engineering", + "kind": "course", + "offer_id": "full-stack-ai-engineering", + "entity_id": "offer:full-stack-ai-engineering", + "title": "Full Stack AI Engineering · Build Real Products with LLMs, Context Engineering, RAG | Towards AI Academy", + "review_title": "Full Stack AI Engineering", + "reviewed_summary": "Production LLM product engineering course.", + "meta_description": "Become a Full Stack AI Engineer. Our flagship 60-hour, project-based program covers prompting, advanced RAG, fine-tuning, agents, and production deployment. Lifetime access for $349.", + "headings": [ + "Become a Full Stack AI Engineer", + "Industry leaders on the course", + "The gaps between developer and AI engineer", + "Learn to engineer around probabilistic systems", + "Get trained in exactly what the role demands", + "Ship real, deployable LLM products", + "Experience the course before you commit", + "What you'll build", + "RAG AI Tutor", + "Full LLM Toolkit", + "Your Own LLM Product", + "Build your first LLM product this week", + "Hear from the engineers who took the leap", + "Who is this course for?", + "The developer making AI their primary skill set", + "The engineer driving change from the inside", + "The developer with a product idea", + "Course curriculum", + "How it works", + "Self-paced, on your schedule", + "Learn by building", + "Active Discord community", + "Certificate upon completion", + "Lifetime access and updates", + "Money-back guarantee", + "Course prerequisites", + "Build your first scalable product", + "Frequently asked questions", + "Ready to build real products with LLMs?" + ], + "text": "Flagship certification · Self-paced\n\nBecome a Full Stack AI Engineer\nAnd position yourself for the roles, projects, and opportunities that AI is creating right now. Try 6 lessons free Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO · co-author, Building LLMs for Production 92 lessons across the full AI stack 60+ hrs of code-heavy, project-based learning 1,000+ AI engineers already enrolled Pay once 3 monthly payments $399 $349 one-time Save 12% Pay once and save $50 versus $399 on the 3-month plan. $133 /month for 3 months $399 in total, with the same lifetime access. Build LLM products end to end: prompting, advanced RAG, fine-tuning, agents Format 92 lessons, 60+ hours, code-heavy and project-based Outcome Certification, a reviewed capstone, and lifetime access with weekly updates Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 Trusted voices\n\nIndustry leaders on the course\n02 Why this course exists\n\nThe gaps between developer and AI engineer\nWe close them with real projects and the exact skills companies are hiring for. Software engineering is the easy half\n\nLearn to engineer around probabilistic systems\nDesign reliable systems on top of unpredictable models, using prompt engineering, context engineering, and RAG to control what the model knows, and fine-tuning to control how it behaves. The talent gap is your opportunity\n\nGet trained in exactly what the role demands\nThe AI engineering role is only 2 to 3 years old; the window to get in early is still open. Master the full stack, from choosing the right models and architectures to evaluating, optimizing, and deploying at scale. Demos are easy. Production is the test.\n\nShip real, deployable LLM products\nMove beyond notebooks and build something that gives you a competitive edge in the job market, launches your startup idea, or drives a new product at your company.\n\nExperience the course before you commit\nExplore the first 6 lessons free. No card required. Unlock free preview 03 Hands-on from day one\n\nWhat you'll build\nA production-grade AI tutor and a complete LLM toolkit through hands-on capstone projects.\n\nRAG AI Tutor\nA fully deployed AI tutor, built from scratch across 60+ lessons. Prompt engineering system prompts and iterative evaluation Data pipeline scrape, parse, and clean data with Firecrawl, LlamaParse, and web APIs RAG pipeline built from scratch, then scaled with LlamaIndex and vector databases Advanced retrieval hybrid search, re-ranking, metadata filtering, query augmentation Fine-tuning GPT-4o Mini and embedding models, optimized for your use case Production deployment FastAPI, Gradio UI, Hugging Face Spaces\n\nFull LLM Toolkit\nThe broader skills the role demands, beyond the central project. Agent architecture tool use, LLM pipelines, autonomous task execution Framework fluency LangChain, LlamaIndex, and OpenAI Assistants side by side Multi-modal capabilities image generation, speech, diffusion models Observability production tracing with Langsmith, Phoenix, and Langfuse LLM optimization quantization, pruning, distillation, speculative decoding\n\nYour Own LLM Product\nEverything applied to an LLM application you choose and ship. Product strategy 175 project ideas, niche selection, defensibility frameworks Business context deep dives on Perplexity, Harvey AI, and Consensus Custom RAG application your own idea, fully built and deployed on Hugging Face Certified delivery reviewed by our team; portfolio-ready for jobs, MVPs, or pitches Incubation opportunity standout projects get direct support to find work or launch\n\nBuild your first LLM product this week\nA fully deployed LLM application / Portfolio-ready skills with certification / Fundamentals that outlast trends Enroll now self-paced · regular updates · risk-free enrollment 04 Student outcomes\n\nHear from the engineers who took the leap\n05 Made for builders\n\nWho is this course for?\nSoftware developers, ML engineers, data scientists, and CS or AI students ready to move into an LLM developer role and start building. Going all-in on AI\n\nThe developer making AI their primary skill set\nBuild a portfolio, earn certification, and walk into high-demand LLM and AI engineering roles with proof you can deliver. Leading AI at their company\n\nThe engineer driving change from the inside\nYou see AI transforming your industry. Gain the hands-on experience to lead initiatives, deploy real solutions, and drive innovation from the inside. Building a product\n\nThe developer with a product idea\nYou have an idea and the technical chops to build it. Turn it into a deployed, production-grade LLM application within 50+ hours of focused learning. 06 The live syllabus\n\nCourse curriculum\nFour parts, 92 lessons, one fully deployed product. The first six lessons are free to preview. 01 Course Introduction and Understanding LLMs 9 lessons + 1 From Beginner to Advanced LLM Developer | The Towards AI Academy Free preview 2 Course Introduction and Logistics Free preview 3 Syllabus In-Depth Free preview 4 How to use LLMs via API Free preview 5 An LLM Developer's Role is to Address AI's Weaknesses and Increase its Capabilities Free preview 6 Why Prompt Engineering, RAG, Tools, and Fine-tuning? Free preview 7 Understanding AI and Large Language Models without Math or Code 8 Alien Intelligence Leads to Unintuitive Mistakes 9 Course Tools Guide 02 Part 1: Building our RAG AI Tutor — Using LLMs 11 lessons + 1 Section Overview: Building Our RAG AI Tutor; Introduction to Using LLMs 2 To use a LLM or to not use it? 3 Choosing your LLM Part 1: Open vs Closed Source and Key Comparison Metrics 4 Choosing your LLM Part 2: Key Benchmarks and Leaderboards 5 Choosing your LLM Part 3: The Leading Open and Closed LLMs Today 6 What is Prompting? Talking with AI Models… 7 Prompting 101 8 What is Prompt Injection? Can you Hack a Prompt? 9 Building our AI Tutor System Prompt. Introducing Prompt Injection and Hacking 10 Evaluating and Iterating Prompts 11 Quiz 03 Part 1: RAG Basics 7 lessons + 1 Section Overview: Using Basic RAG for Our Project 2 What is RAG? 3 RAG 101 4 Building your Knowledge Base for RAG 5 Building a basic RAG pipeline from scratch 6 Build an OpenAI GPT with our AI blog data 7 Quiz 04 Part 1: RAG with LlamaIndex 8 lessons + 1 Section Overview: Developing a RAG AI Tutor With LlamaIndex 2 Basic RAG with LlamaIndex 3 How vector DBs work and when to use one 4 Using a Vector Database 5 Improving Data Sources and Prompts 6 RAG Evaluations 7 Evaluating your RAG Pipeline 8 Quiz 05 Part 1: Using other LLMs and Embedding Models 7 lessons + 1 Section Overview: Using Other LLMs and Embedding Models 2 Security, Privacy & Cost; Comparing 5 key ways to access LLMs 3 Using Llama 3.1 70B on Together.ai and comparing RAG performance 4 Ollama Tutorial: Running Deepseek Distill Locally 5 Multimodal LLMs and Their Role in RAG Pipelines 6 How to Select the Right Embedding Model for Your Use Case? 7 Quiz 06 Part 1: Collecting, Filtering and Cleaning Data 9 lessons + 1 Section Overview: Collecting, Filtering, and Cleaning Data for RAG and LLM Pipelines 2 Scraping Data from Websites 3 Scraping via API with Firecrawl 4 Perplexity Web API 5 Web Search API 6 Structuring Your Data: OpenAI Structured JSON Outputs 7 Parsing Data with LlamaParse 8 Choosing Data Sources and Building our AI Tutor Dataset 9 Quiz 07 Part 1: Advanced RAG 11 lessons + 1 Section Overview: Advanced RAG; How we Find and Use the Most Relevant Data 2 Larger Context Window; More Chunks 3 LLM as a judge; Ranking Chunks 4 Search with Metadata Filtering 5 Advanced Search Techniques: From Keywords to Graphs 6 Hybrid search 7 Query Variation and Augmentation 8 Re-Ranking 9 What is Indexing? Indexing Methods for Vector Retrieval 10 Advanced Retrieval 11 Quiz 08 Part 1: Fine-Tuning 7 lessons + 1 Section Overview: Fine-Tuning 2 Optimizing the model: inference and fine-tuning 3 Fine-tuning 101 4 GPT4o-Mini Fine Tuning in our project 5 Is fine-tuning an embedding model worth it? When should you do it? 6 Fine-Tuning Embedding Models with LlamaIndex 7 Quiz 09 Part 1: Expanded RAG Toolkit 11 lessons + 1 Section Overview: Expanded RAG Toolkit 2 Non-Technical Skills for Effective LLM Development 3 Adding Question Validation and Routing 4 An Introduction to Reasoning Language Models 5 Reinforcement Fine-Tuning for OpenAI 6 Enabling Conversational Memory in LLMs 7 Adding Speech with GPT-4o Audio 8 Long Context LLMs vs. RAG 9 Long Context LLMs & Context Caching vs RAG 10 Graph ML and GraphRAG 11 Quiz 10 Part 1: APIs, UI and Deployment 6 lessons + 1 Section Overview: Building and Deploying the Final RAG Chatbot 2 Bringing it all together: Building the final AI Tutor 3 How to make an API? 4 Building and Deploying a Gradio UI on Hugging Face Spaces 5 Quick Tips for Improving LLM Apps Once in Production 6 Quiz 11 Part 2: More LLM Capabilities and Useful AI Models 6 lessons + 1 Part 2 Intro; Introducing more tools, models & frameworks for LLM projects 2 Section Overview: More LLM Capabilities & Other Useful AI Models 3 Image Generation (Diffusion) and Speech models 4 Generating and Editing Images via the OpenAI API 5 LLM Agents: LLM Pipelines and Tool Use for Complex Tasks 6 Quiz 12 Part 2: More LLM Frameworks and Tools 10 lessons + 1 Section Overview: More LLM Frameworks and Tools 2 More LLM framework options; LangChain vs. LlamaIndex vs. OpenAI Assistants 3 LangChain Introduction 101 4 Access More AI Models with Hugging Face's APIs 5 Building Agents with OpenAI Assistants 6 LangChain OpenGPT 7 Extending LLM Capabilities with Anthropic's Model Context Protocol (MCP) 8 Observability or tracing in production; Langsmith, Phoenix and Langfuse 9 More APIs and tools 10 Quiz 13 Part 2: LLM Optimization 7 lessons + 1 Section Overview: LLM Optimizations for Deployment 2 Best tips for Pruning and Distillation (Minitron Paper) 3 Model Distillation and Teacher-Student Models 4 LLM Deployment Optimisation; Quantization, Pruning, Speculative Decoding 5 Deploying a Quantized LLM on CPU on Google Cloud Platform (GCP) 6 Deploying an LLM Using Latitude.sh 7 Quiz 14 Part 3: LLM Ecosystem Economics; Choosing Your Niche and Business Strategy 10 lessons + 1 Part 3 Intro (optional); Deciding which LLM RAG project to build yourself 2 What will drive Enterprise adoption of Generative AI? 3 Who are the Generative AI Winners? How will Adoption Reshape the Economy? 4 Impact of Large Language Models on Different Job Categories 5 175 LLM Project ideas; Choosing your industry, niche, primary function and target users 6 What Makes a Defensible Product Built with Foundation LLMs? 7 AI LLM/RAG deep dive; Perplexity 8 LLM/RAG deep dive; Consensus 9 AI/LLM Deep dive; Harvey AI 10 Quiz 15 Part 4: Project + Extra Tools 5 lessons + 1 Building Your Own advanced LLM + RAG Project to receive certification 2 Project Submission 3 Share your success and showcase your new skills! 4 AI Tutor: Ask any question! 5 Staying Ahead of AI Developments Updated every two weeks with new lessons as the field evolves. Pay once 3 monthly payments $399 $349 one-time Save 12% Pay once and save $50 versus $399 on the 3-month plan. $133 /month for 3 months $399 in total, with the same lifetime access. Build LLM products end to end: prompting, advanced RAG, fine-tuning, agents Format 92 lessons, 60+ hours, code-heavy and project-based Outcome Certification, a reviewed capstone, and lifetime access with weekly updates Discord community and AI tutor support Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 07 The learning experience\n\nHow it works\nProject-based learning designed for working professionals, fully self-paced. 01\n\nSelf-paced, on your schedule\nStart the day you enroll, move at your own pace, and revisit any lesson when a project demands it. 02\n\nLearn by building\nHands-on projects from day one. Build a production LLM system from scratch rather than following along with toy examples. 03\n\nActive Discord community\nGet unstuck fast. Connect with fellow builders, share progress, and collaborate on projects in an active engineering community. 04\n\nCertificate upon completion\nGraduate with certification backed by a real deployed system: portfolio-ready proof you ship. 05\n\nLifetime access and updates\nAll current content plus future updates as AI evolves. Your investment stays valuable. 06\n\nMoney-back guarantee\n30-day full refund, no questions asked. Risk-free enrollment. 08 Before you start\n\nCourse prerequisites\nThe projects are hands-on, so a technical foundation matters. 01 Intermediate Python knowledge 02 Basic knowledge of GitHub 03 A device you can set up a coding environment on 04 At least 50 hours to gain the full benefit of the course New to Python? Start with our Beginner Python for AI Engineering course first. Enrollment\n\nBuild your first scalable product\nWith LLMs, prompting, RAG, tool use, and fine-tuning. $349 one-time, with lifetime access and all future updates. Unlock access for $349 Secure checkout via Towards AI Academy. 30-day money-back guarantee: if you explore the early lessons and it's not what you need, email us for a full refund. 09 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What projects will I work on? + You'll build your own AI tutor, create RAG pipelines, and ship an advanced LLM MVP. Each project is designed to solve real-world problems, so you graduate with practical, hands-on experience. What are the course prerequisites? + Intermediate Python, basic GitHub knowledge, and a device you can set up a coding environment on. No prior AI or LLM experience needed. When was this course created? Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field. How is this course different from other AI courses? + It is fully hands-on, project-based, and focused on the skills top companies are hiring for right now. We guide you through every step, so you understand the concepts and can apply them immediately. How does this course compare to your \"Agent Engineering\" course? + Full Stack AI Engineering covers the complete LLM product lifecycle: data collection, RAG, fine-tuning, deployment, and business positioning, with agents as one component. The Agents course goes much deeper into designing, orchestrating, evaluating, and deploying autonomous systems. Compare the Agent Engineering course . Do you have a referral or affiliate program? + Yes. You can earn cashback, free courses, and up to $70 per course or $180 per bundle in commission by promoting our courses and the \"Building LLMs for Production\" eBook. See the Affiliate page for tiers and payouts. Can I get a discount? + Yes, there are three ways to receive a discount. Students get 50% off all courses. If you've previously purchased our book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Can I take this course within my company? + Yes. We offer both course bundles and custom training solutions tailored specifically for companies. For company packages or a customized training plan, reach out to Louis at louis@towardsai.net . Lifetime access · $349\n\nReady to build real products with LLMs?\nJoin 1,000+ AI engineers. Self-paced, updated weekly, and backed by a 30-day money-back guarantee. Unlock lifetime access for $349 Try 6 lessons free", + "links": [ + { + "text": "Try 6 lessons free", + "url": "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/" + }, + { + "text": "Unlock lifetime access", + "url": "https://academy.towardsai.net/cart/add_product/2858239?price_id=3697529" + }, + { + "text": "Start the 3-month plan", + "url": "https://academy.towardsai.net/enroll/2858239?price_id=4512010" + }, + { + "text": "Unlock free preview", + "url": "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/" + }, + { + "text": "Enroll now", + "url": "https://academy.towardsai.net/cart/add_product/2858239?price_id=3697529" + }, + { + "text": "From Beginner to Advanced LLM Developer | The Towards AI Academy Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "Course Introduction and Logistics Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "Syllabus In-Depth Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "How to use LLMs via API Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "An LLM Developer's Role is to Address AI's Weaknesses and Increase its Capabilities Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "Why Prompt Engineering, RAG, Tools, and Fine-tuning? Free preview", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + }, + { + "text": "Beginner Python for AI Engineering", + "url": "https://towardsai.com/academy/python-for-ai-engineering/" + }, + { + "text": "Unlock access for $349", + "url": "https://academy.towardsai.net/cart/add_product/2858239?price_id=3697529" + }, + { + "text": "Compare the Agent Engineering course", + "url": "https://towardsai.com/academy/agent-engineering/" + }, + { + "text": "Affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + }, + { + "text": "Unlock lifetime access for $349", + "url": "https://academy.towardsai.net/cart/add_product/2858239?price_id=3697529" + } + ], + "chunks": [ + { + "chunk_id": "tai-cd07542bb8452f94c8d1621e", + "text": "Flagship certification · Self-paced\n\nBecome a Full Stack AI Engineer\nAnd position yourself for the roles, projects, and opportunities that AI is creating right now. Try 6 lessons free Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO · co-author, Building LLMs for Production 92 lessons across the full AI stack 60+ hrs of code-heavy, project-based learning 1,000+ AI engineers already enrolled Pay once 3 monthly payments $399 $349 one-time Save 12% Pay once and save $50 versus $399 on the 3-month plan. $133 /month for 3 months $399 in total, with the same lifetime access. Build LLM products end to end: prompting, advanced RAG, fine-tuning, agents Format 92 lessons, 60+ hours, code-heavy and project-based Outcome Certification, a reviewed capstone, and lifetime access with weekly updates Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 Trusted voices\n\nIndustry leaders on the course\n02 Why this course exists\n\nThe gaps between developer and AI engineer\nWe close them with real projects and the exact skills companies are hiring for. Software engineering is the easy half\n\nLearn to engineer around probabilistic systems\nDesign reliable systems on top of unpredictable models, using prompt engineering, context engineering, and RAG to control what the model knows, and fine-tuning to control how it behaves. The talent gap is your opportunity\n\nGet trained in exactly what the role demands\nThe AI engineering role is only 2 to 3 years old; the window to get in early is still open. Master the full stack, from choosing the right models and architectures to evaluating, optimizing, and deploying at scale. Demos are easy. Production is the test.", + "heading": "Become a Full Stack AI Engineer", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-e80906d03bde3a1cda798854", + "text": "Flagship certification · Self-paced" + }, + { + "span_id": "span-81e592df81183970c99587cb", + "text": "And position yourself for the roles, projects, and opportunities that AI is creating right now." + }, + { + "span_id": "span-d0f53effcca37d23add67ea0", + "text": "Try 6 lessons free" + }, + { + "span_id": "span-55361c09f6e4825b386f0d13", + "text": "Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO · co-author, Building LLMs for Production" + }, + { + "span_id": "span-55f473a7b537722322a95f40", + "text": "92 lessons across the full AI stack" + }, + { + "span_id": "span-93e8727641dd4615f2037286", + "text": "60+ hrs of code-heavy, project-based learning" + }, + { + "span_id": "span-ad4dc4704e0339c10287444a", + "text": "1,000+ AI engineers already enrolled" + }, + { + "span_id": "span-9369c120374f0407d1eabe1e", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-53929fa16414dfbd6e34bfb9", + "text": "$399 $349 one-time Save 12%" + }, + { + "span_id": "span-d613ee9035162d940b8e717f", + "text": "Pay once and save $50 versus $399 on the 3-month plan." + }, + { + "span_id": "span-c5c2c6de19717ce2ba655265", + "text": "$133 /month for 3 months" + }, + { + "span_id": "span-731d87f7b312d4c0a01688f4", + "text": "$399 in total, with the same lifetime access." + }, + { + "span_id": "span-360518ffead107417c2cbff6", + "text": "LLM products end to end: prompting, advanced RAG, fine-tuning, agents" + }, + { + "span_id": "span-dcec419321df7320b4333804", + "text": "92 lessons, 60+ hours, code-heavy and project-based" + }, + { + "span_id": "span-acb2d38a62ef5b1dc84655b0", + "text": "Certification, a reviewed capstone, and lifetime access with weekly updates" + }, + { + "span_id": "span-ad6c6c9339d7e5ce5c27f550", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-1b3f99c87df3fcc4d1f758ab", + "text": "01 Trusted voices" + }, + { + "span_id": "span-97af157520dbed6cf753be35", + "text": "02 Why this course exists" + }, + { + "span_id": "span-2d9e74df60f022cd19a8479b", + "text": "We close them with real projects and the exact skills companies are hiring for." + }, + { + "span_id": "span-8e4cb3fc7a4cf097e91b6fe4", + "text": "Software engineering is the easy half" + }, + { + "span_id": "span-9a39e560558b9534d71f5486", + "text": "Design reliable systems on top of unpredictable models, using prompt engineering, context engineering, and RAG to control what the model knows, and fine-tuning to control how it behaves." + }, + { + "span_id": "span-1213b6fc66ee850721ae76f0", + "text": "The talent gap is your opportunity" + }, + { + "span_id": "span-2cdb4ecabcc6a667c02b9f5b", + "text": "The AI engineering role is only 2 to 3 years old; the window to get in early is still open." + }, + { + "span_id": "span-bc64804e6138a27c5f6b57df", + "text": "Master the full stack, from choosing the right models and architectures to evaluating, optimizing, and deploying at scale." + }, + { + "span_id": "span-7a74dc0eaf8e01b56f1fecda", + "text": "Demos are easy." + }, + { + "span_id": "span-67d8fc1cae96e775a8e871ae", + "text": "Production is the test." + } + ] + }, + { + "chunk_id": "tai-41566976411a73d934345011", + "text": "Ship real, deployable LLM products\nMove beyond notebooks and build something that gives you a competitive edge in the job market, launches your startup idea, or drives a new product at your company.\n\nExperience the course before you commit\nExplore the first 6 lessons free. No card required. Unlock free preview 03 Hands-on from day one\n\nWhat you'll build\nA production-grade AI tutor and a complete LLM toolkit through hands-on capstone projects.\n\nRAG AI Tutor\nA fully deployed AI tutor, built from scratch across 60+ lessons. Prompt engineering system prompts and iterative evaluation Data pipeline scrape, parse, and clean data with Firecrawl, LlamaParse, and web APIs RAG pipeline built from scratch, then scaled with LlamaIndex and vector databases Advanced retrieval hybrid search, re-ranking, metadata filtering, query augmentation Fine-tuning GPT-4o Mini and embedding models, optimized for your use case Production deployment FastAPI, Gradio UI, Hugging Face Spaces\n\nFull LLM Toolkit\nThe broader skills the role demands, beyond the central project. Agent architecture tool use, LLM pipelines, autonomous task execution Framework fluency LangChain, LlamaIndex, and OpenAI Assistants side by side Multi-modal capabilities image generation, speech, diffusion models Observability production tracing with Langsmith, Phoenix, and Langfuse LLM optimization quantization, pruning, distillation, speculative decoding", + "heading": "RAG AI Tutor", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-d22e9a24274b3b1217a12ce6", + "text": "Move beyond notebooks and build something that gives you a competitive edge in the job market, launches your startup idea, or drives a new product at your company." + }, + { + "span_id": "span-04fd8621f856b07b4d4f91a6", + "text": "Explore the first 6 lessons free." + }, + { + "span_id": "span-05f528003e804f2a4bdf0966", + "text": "No card required." + }, + { + "span_id": "span-e16adee3e82c0097ff07931e", + "text": "Unlock free preview" + }, + { + "span_id": "span-9f01b22075107be1224a328a", + "text": "03 Hands-on from day one" + }, + { + "span_id": "span-a963752aff62039021017f69", + "text": "A production-grade AI tutor and a complete LLM toolkit through hands-on capstone projects." + }, + { + "span_id": "span-e1604aa0f945c8614cc1e076", + "text": "A fully deployed AI tutor, built from scratch across 60+ lessons." + }, + { + "span_id": "span-b7bcf2efbedaf72ea6dc9114", + "text": "Prompt engineering system prompts and iterative evaluation" + }, + { + "span_id": "span-dd1c42c3d143c6ee2bc4b791", + "text": "Data pipeline scrape, parse, and clean data with Firecrawl, LlamaParse, and web APIs" + }, + { + "span_id": "span-b14beb0566100dda55d219e1", + "text": "RAG pipeline built from scratch, then scaled with LlamaIndex and vector databases" + }, + { + "span_id": "span-41fb4aec0d5d511bb7799be4", + "text": "Advanced retrieval hybrid search, re-ranking, metadata filtering, query augmentation" + }, + { + "span_id": "span-2a2bd6e0064a60aa835b3a94", + "text": "Fine-tuning GPT-4o Mini and embedding models, optimized for your use case" + }, + { + "span_id": "span-dba40e9f5c3b546019104ad3", + "text": "Production deployment FastAPI, Gradio UI, Hugging Face Spaces" + }, + { + "span_id": "span-3cc0a69df2a22571e8af94e2", + "text": "The broader skills the role demands, beyond the central project." + }, + { + "span_id": "span-903cc8eb78d18cb8bd322851", + "text": "Agent architecture tool use, LLM pipelines, autonomous task execution" + }, + { + "span_id": "span-8025d65665bb3784effdb6ac", + "text": "Framework fluency LangChain, LlamaIndex, and OpenAI Assistants side by side" + }, + { + "span_id": "span-a5387435d3966422e6d1edf2", + "text": "Multi-modal capabilities image generation, speech, diffusion models" + }, + { + "span_id": "span-5d87e84487ab23b3fb1dacb8", + "text": "Observability production tracing with Langsmith, Phoenix, and Langfuse" + }, + { + "span_id": "span-ac290e8d1c9facf14e181100", + "text": "LLM optimization quantization, pruning, distillation, speculative decoding" + } + ] + }, + { + "chunk_id": "tai-814b1e6dd32c8630324777a4", + "text": "Your Own LLM Product\nEverything applied to an LLM application you choose and ship. Product strategy 175 project ideas, niche selection, defensibility frameworks Business context deep dives on Perplexity, Harvey AI, and Consensus Custom RAG application your own idea, fully built and deployed on Hugging Face Certified delivery reviewed by our team; portfolio-ready for jobs, MVPs, or pitches Incubation opportunity standout projects get direct support to find work or launch\n\nBuild your first LLM product this week\nA fully deployed LLM application / Portfolio-ready skills with certification / Fundamentals that outlast trends Enroll now self-paced · regular updates · risk-free enrollment 04 Student outcomes\n\nHear from the engineers who took the leap\n05 Made for builders\n\nWho is this course for?\nSoftware developers, ML engineers, data scientists, and CS or AI students ready to move into an LLM developer role and start building. Going all-in on AI\n\nThe developer making AI their primary skill set\nBuild a portfolio, earn certification, and walk into high-demand LLM and AI engineering roles with proof you can deliver. Leading AI at their company\n\nThe engineer driving change from the inside\nYou see AI transforming your industry. Gain the hands-on experience to lead initiatives, deploy real solutions, and drive innovation from the inside. Building a product\n\nThe developer with a product idea\nYou have an idea and the technical chops to build it. Turn it into a deployed, production-grade LLM application within 50+ hours of focused learning. 06 The live syllabus", + "heading": "Your Own LLM Product", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-3c6e85a3ad817f59702026cd", + "text": "Everything applied to an LLM application you choose and ship." + }, + { + "span_id": "span-5320aa9428104157619860e5", + "text": "Product strategy 175 project ideas, niche selection, defensibility frameworks" + }, + { + "span_id": "span-516f366799f878106c6afaa4", + "text": "Business context deep dives on Perplexity, Harvey AI, and Consensus" + }, + { + "span_id": "span-9125442a9684cabac058f0d3", + "text": "Custom RAG application your own idea, fully built and deployed on Hugging Face" + }, + { + "span_id": "span-8f30d6ad84bde87542ca7560", + "text": "Certified delivery reviewed by our team; portfolio-ready for jobs, MVPs, or pitches" + }, + { + "span_id": "span-e5ed831987d0e81bf2183605", + "text": "Incubation opportunity standout projects get direct support to find work or launch" + }, + { + "span_id": "span-43bcfab55da1d609af93129e", + "text": "A fully deployed LLM application / Portfolio-ready skills with certification / Fundamentals that outlast trends" + }, + { + "span_id": "span-900b1e1e467681fba2a66c7e", + "text": "Enroll now self-paced · regular updates · risk-free enrollment" + }, + { + "span_id": "span-b2438d9a8967080ac1335059", + "text": "04 Student outcomes" + }, + { + "span_id": "span-9e876fc3a5f4aadcc15dbdc9", + "text": "05 Made for builders" + }, + { + "span_id": "span-33f6f585aeb9ca18b927ba65", + "text": "Software developers, ML engineers, data scientists, and CS or AI students ready to move into an LLM developer role and start building." + }, + { + "span_id": "span-71cd77f52d31ea6500dfc920", + "text": "Going all-in on AI" + }, + { + "span_id": "span-3796cd76e2cc3b5592d4b57a", + "text": "Build a portfolio, earn certification, and walk into high-demand LLM and AI engineering roles with proof you can deliver." + }, + { + "span_id": "span-791ea9daf4b85e45afaf676a", + "text": "Leading AI at their company" + }, + { + "span_id": "span-1104fb6762a8209c600f5104", + "text": "You see AI transforming your industry." + }, + { + "span_id": "span-464b0f0d2b7dc95c1d995b4f", + "text": "Gain the hands-on experience to lead initiatives, deploy real solutions, and drive innovation from the inside." + }, + { + "span_id": "span-99fec4859a4494e2e0c51ec9", + "text": "Building a product" + }, + { + "span_id": "span-701d440daaae79c2741d7ccf", + "text": "You have an idea and the technical chops to build it." + }, + { + "span_id": "span-384144911c2b54cf5add9346", + "text": "Turn it into a deployed, production-grade LLM application within 50+ hours of focused learning." + }, + { + "span_id": "span-961a9c58bbfe4d122c7a8c2b", + "text": "06 The live syllabus" + } + ] + }, + { + "chunk_id": "tai-26d5c3d318796605e692f771", + "text": "Course curriculum\nFour parts, 92 lessons, one fully deployed product. The first six lessons are free to preview. 01 Course Introduction and Understanding LLMs 9 lessons + 1 From Beginner to Advanced LLM Developer | The Towards AI Academy Free preview 2 Course Introduction and Logistics Free preview 3 Syllabus In-Depth Free preview 4 How to use LLMs via API Free preview 5 An LLM Developer's Role is to Address AI's Weaknesses and Increase its Capabilities Free preview 6 Why Prompt Engineering, RAG, Tools, and Fine-tuning? Free preview 7 Understanding AI and Large Language Models without Math or Code 8 Alien Intelligence Leads to Unintuitive Mistakes 9 Course Tools Guide 02 Part 1: Building our RAG AI Tutor — Using LLMs 11 lessons + 1 Section Overview: Building Our RAG AI Tutor; Introduction to Using LLMs 2 To use a LLM or to not use it? 3 Choosing your LLM Part 1: Open vs Closed Source and Key Comparison Metrics 4 Choosing your LLM Part 2: Key Benchmarks and Leaderboards 5 Choosing your LLM Part 3: The Leading Open and Closed LLMs Today 6 What is Prompting? Talking with AI Models… 7 Prompting 101 8 What is Prompt Injection? Can you Hack a Prompt? 9 Building our AI Tutor System Prompt. Introducing Prompt Injection and Hacking 10 Evaluating and Iterating Prompts 11 Quiz 03 Part 1: RAG Basics 7 lessons + 1 Section Overview: Using Basic RAG for Our Project 2 What is RAG? 3 RAG 101 4 Building your Knowledge Base for RAG 5 Building a basic RAG pipeline from scratch 6 Build an OpenAI GPT with our AI blog data 7 Quiz 04 Part 1: RAG with LlamaIndex 8 lessons + 1 Section Overview: Developing a RAG AI Tutor With LlamaIndex 2 Basic RAG with LlamaIndex 3 How vector DBs work and when to use one 4 Using a Vector Database 5 Improving Data Sources and Prompts 6 RAG Evaluations 7", + "heading": "Course curriculum", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-9cd00424bcf54d3baeb6912e", + "text": "Four parts, 92 lessons, one fully deployed product." + }, + { + "span_id": "span-922c06dc3270b1c32aeca157", + "text": "The first six lessons are free to preview." + }, + { + "span_id": "span-f64633d4b9472e144fc2a228", + "text": "01 Course Introduction and Understanding LLMs 9 lessons + 1 From Beginner to Advanced LLM Developer | The Towards AI Academy Free preview" + }, + { + "span_id": "span-38190d5254b8a81d2f36a990", + "text": "2 Course Introduction and Logistics Free preview" + }, + { + "span_id": "span-0fdbf0b7caebb0ac893bca2e", + "text": "3 Syllabus In-Depth Free preview" + }, + { + "span_id": "span-3c20a34abe257dee57a6d1db", + "text": "4 How to use LLMs via API Free preview" + }, + { + "span_id": "span-30b58510369eb163c3a7bdd8", + "text": "5 An LLM Developer's Role is to Address AI's Weaknesses and Increase its Capabilities Free preview" + }, + { + "span_id": "span-11314f78ef6870cf3e0520f5", + "text": "6 Why Prompt Engineering, RAG, Tools, and Fine-tuning?" + }, + { + "span_id": "span-3eb5d4dee5ea819090b6f741", + "text": "Free preview" + }, + { + "span_id": "span-63f628efd095e20c575562cf", + "text": "7 Understanding AI and Large Language Models without Math or Code" + }, + { + "span_id": "span-11e5657cb12a1509fda3e338", + "text": "8 Alien Intelligence Leads to Unintuitive Mistakes" + }, + { + "span_id": "span-37934fb3f47593e67fff8e66", + "text": "9 Course Tools Guide" + }, + { + "span_id": "span-c12d936962feadd7728b18c0", + "text": "02 Part 1: Building our RAG AI Tutor — Using LLMs 11 lessons + 1 Section Overview: Building Our RAG AI Tutor; Introduction to Using LLMs" + }, + { + "span_id": "span-de7c01ba6345047d0e44a73d", + "text": "2 To use a LLM or to not use it?" + }, + { + "span_id": "span-91dd4480b991570174ff8145", + "text": "3 Choosing your LLM Part 1: Open vs Closed Source and Key Comparison Metrics" + }, + { + "span_id": "span-bc4e69d9147c0522228ad7d0", + "text": "4 Choosing your LLM Part 2: Key Benchmarks and Leaderboards" + }, + { + "span_id": "span-7ab1fe7d8b93fb88f4b88837", + "text": "5 Choosing your LLM Part 3: The Leading Open and Closed LLMs Today" + }, + { + "span_id": "span-5bd6d2339e93eab580cc4311", + "text": "6 What is Prompting?" + }, + { + "span_id": "span-e165fe2b708e274aa6e5e594", + "text": "Talking with AI Models…" + }, + { + "span_id": "span-6b634d2f3e3f214edd48f5f9", + "text": "7 Prompting 101" + }, + { + "span_id": "span-837119dbc6fbd16210a5d93d", + "text": "8 What is Prompt Injection?" + }, + { + "span_id": "span-285d65abcebcc26b4d182210", + "text": "Can you Hack a Prompt?" + }, + { + "span_id": "span-df90c6e6cb67980fe9b13fef", + "text": "9 Building our AI Tutor System Prompt." + }, + { + "span_id": "span-dfee05d9d5990c90f690cb2b", + "text": "Introducing Prompt Injection and Hacking" + }, + { + "span_id": "span-4dd00bc88b0d0024d8764147", + "text": "10 Evaluating and Iterating Prompts" + }, + { + "span_id": "span-76d30d7d65e3d79e4d164bf3", + "text": "11 Quiz" + }, + { + "span_id": "span-9d34e2a27dbc021e0abb7256", + "text": "03 Part 1: RAG Basics 7 lessons + 1 Section Overview: Using Basic RAG for Our Project" + }, + { + "span_id": "span-5e12f2688b289264689d17bf", + "text": "2 What is RAG?" + }, + { + "span_id": "span-3a4b4c14e4dd05d08d1e41e6", + "text": "3 RAG 101" + }, + { + "span_id": "span-5eb9dcca0162c3ed610894a3", + "text": "4 Building your Knowledge Base for RAG" + }, + { + "span_id": "span-7992fe4ba4d8258035b5a33a", + "text": "5 Building a basic RAG pipeline from scratch" + }, + { + "span_id": "span-b35c1bf96c2cdc6a618e0f6d", + "text": "6 Build an OpenAI GPT with our AI blog data" + }, + { + "span_id": "span-dcfb441f8d4199a641a08732", + "text": "7 Quiz" + }, + { + "span_id": "span-caaef7e30eb36a3f2f83f932", + "text": "04 Part 1: RAG with LlamaIndex 8 lessons + 1 Section Overview: Developing a RAG AI Tutor With LlamaIndex" + }, + { + "span_id": "span-bceccd75019bc555c991fff5", + "text": "2 Basic RAG with LlamaIndex" + }, + { + "span_id": "span-89084d599ebbf1b9cbd7138b", + "text": "3 How vector DBs work and when to use one" + }, + { + "span_id": "span-f0a468b20cd38b22a25809fa", + "text": "4 Using a Vector Database" + }, + { + "span_id": "span-234c780306548f30d6599e91", + "text": "5 Improving Data Sources and Prompts" + }, + { + "span_id": "span-9e20b68abe8dd3e2fa5395da", + "text": "6 RAG Evaluations" + } + ] + }, + { + "chunk_id": "tai-a86a9c5ea6b401beafe0228b", + "text": "Course curriculum\nEvaluating your RAG Pipeline 8 Quiz 05 Part 1: Using other LLMs and Embedding Models 7 lessons + 1 Section Overview: Using Other LLMs and Embedding Models 2 Security, Privacy & Cost; Comparing 5 key ways to access LLMs 3 Using Llama 3.1 70B on Together.ai and comparing RAG performance 4 Ollama Tutorial: Running Deepseek Distill Locally 5 Multimodal LLMs and Their Role in RAG Pipelines 6 How to Select the Right Embedding Model for Your Use Case? 7 Quiz 06 Part 1: Collecting, Filtering and Cleaning Data 9 lessons + 1 Section Overview: Collecting, Filtering, and Cleaning Data for RAG and LLM Pipelines 2 Scraping Data from Websites 3 Scraping via API with Firecrawl 4 Perplexity Web API 5 Web Search API 6 Structuring Your Data: OpenAI Structured JSON Outputs 7 Parsing Data with LlamaParse 8 Choosing Data Sources and Building our AI Tutor Dataset 9 Quiz 07 Part 1: Advanced RAG 11 lessons + 1 Section Overview: Advanced RAG; How we Find and Use the Most Relevant Data 2 Larger Context Window; More Chunks 3 LLM as a judge; Ranking Chunks 4 Search with Metadata Filtering 5 Advanced Search Techniques: From Keywords to Graphs 6 Hybrid search 7 Query Variation and Augmentation 8 Re-Ranking 9 What is Indexing? Indexing Methods for Vector Retrieval 10 Advanced Retrieval 11 Quiz 08 Part 1: Fine-Tuning 7 lessons + 1 Section Overview: Fine-Tuning 2 Optimizing the model: inference and fine-tuning 3 Fine-tuning 101 4 GPT4o-Mini Fine Tuning in our project 5 Is fine-tuning an embedding model worth it? When should you do it? 6 Fine-Tuning Embedding Models with LlamaIndex 7 Quiz 09 Part 1: Expanded RAG Toolkit 11 lessons + 1 Section Overview: Expanded RAG Toolkit 2 Non-Technical Skills for Effective LLM Development 3 Adding Question Validation and Routing 4 An Introduction to", + "heading": "Course curriculum", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-99faa6c45783a75408689752", + "text": "11 Quiz" + }, + { + "span_id": "span-8df81364a0b008818965b537", + "text": "7 Quiz" + }, + { + "span_id": "span-6334b296cb401ade499ba923", + "text": "8 Quiz" + }, + { + "span_id": "span-9d0fccece5b9d74a6f213a57", + "text": "05 Part 1: Using other LLMs and Embedding Models 7 lessons + 1 Section Overview: Using Other LLMs and Embedding Models" + }, + { + "span_id": "span-42f07a3f4f84b6869887bde0", + "text": "2 Security, Privacy & Cost; Comparing 5 key ways to access LLMs" + }, + { + "span_id": "span-d7376b620c4b1a994deb1aec", + "text": "3 Using Llama 3.1 70B on Together.ai and comparing RAG performance" + }, + { + "span_id": "span-9c0f465fca9d71a4ef3a15cd", + "text": "4 Ollama Tutorial: Running Deepseek Distill Locally" + }, + { + "span_id": "span-6bb668afc4212625e209285c", + "text": "5 Multimodal LLMs and Their Role in RAG Pipelines" + }, + { + "span_id": "span-5eeb50fa2618edf0165df0cd", + "text": "6 How to Select the Right Embedding Model for Your Use Case?" + }, + { + "span_id": "span-4c76b6a76201333607179d23", + "text": "06 Part 1: Collecting, Filtering and Cleaning Data 9 lessons + 1 Section Overview: Collecting, Filtering, and Cleaning Data for RAG and LLM Pipelines" + }, + { + "span_id": "span-abc50d0c71fde0ddb3832863", + "text": "2 Scraping Data from Websites" + }, + { + "span_id": "span-1c16bd2f0f5b462baf04c3f3", + "text": "3 Scraping via API with Firecrawl" + }, + { + "span_id": "span-fccb0613c200cdba54ab53c0", + "text": "4 Perplexity Web API" + }, + { + "span_id": "span-68efb5d157b8463b82fa4d35", + "text": "5 Web Search API" + }, + { + "span_id": "span-0ad8df397a755c485d57af1d", + "text": "6 Structuring Your Data: OpenAI Structured JSON Outputs" + }, + { + "span_id": "span-98e182393da3e2ad45828976", + "text": "7 Parsing Data with LlamaParse" + }, + { + "span_id": "span-2ada41f0c6da4b5088562940", + "text": "8 Choosing Data Sources and Building our AI Tutor Dataset" + }, + { + "span_id": "span-c1bd30d8ba8db9bab336c5a5", + "text": "9 Quiz" + }, + { + "span_id": "span-c37c6c60369a24846a7e7379", + "text": "07 Part 1: Advanced RAG 11 lessons + 1 Section Overview: Advanced RAG; How we Find and Use the Most Relevant Data" + }, + { + "span_id": "span-59c9641a93d23c6eab374050", + "text": "2 Larger Context Window; More Chunks" + }, + { + "span_id": "span-836b58b474f0e757d230dc62", + "text": "3 LLM as a judge; Ranking Chunks" + }, + { + "span_id": "span-dff1343308f459b696ab6174", + "text": "4 Search with Metadata Filtering" + }, + { + "span_id": "span-4d1aa67fa7c8f47716e4843e", + "text": "5 Advanced Search Techniques: From Keywords to Graphs" + }, + { + "span_id": "span-75bd2d21c400ad8abaa8f5b2", + "text": "6 Hybrid search" + }, + { + "span_id": "span-c81dfa7e401b4237b1d05d78", + "text": "7 Query Variation and Augmentation" + }, + { + "span_id": "span-dc95083972cd7d1deeeb1cfa", + "text": "8 Re-Ranking" + }, + { + "span_id": "span-7f1e30ff1fa697b99187aad8", + "text": "9 What is Indexing?" + }, + { + "span_id": "span-f5ca6dadacb3429c54c871ef", + "text": "Indexing Methods for Vector Retrieval" + }, + { + "span_id": "span-fa42569ea2321cdece86010f", + "text": "10 Advanced Retrieval" + }, + { + "span_id": "span-66cda2cc911e2fa16a894726", + "text": "08 Part 1: Fine-Tuning 7 lessons + 1 Section Overview: Fine-Tuning" + }, + { + "span_id": "span-dfa934f6f708c7cca0a6f0fc", + "text": "2 Optimizing the model: inference and fine-tuning" + }, + { + "span_id": "span-63d452520b367e115f32cbdc", + "text": "3 Fine-tuning 101" + }, + { + "span_id": "span-4990437d9d3b1109eca1a1b5", + "text": "4 GPT4o-Mini Fine Tuning in our project" + }, + { + "span_id": "span-50af69dbe22b720bcdad5aea", + "text": "5 Is fine-tuning an embedding model worth it?" + }, + { + "span_id": "span-dc6f502023c082da769d085a", + "text": "When should you do it?" + }, + { + "span_id": "span-5609f4e57c9b182cdfef85c4", + "text": "6 Fine-Tuning Embedding Models with LlamaIndex" + }, + { + "span_id": "span-c2b6508a869b9d150b13490a", + "text": "09 Part 1: Expanded RAG Toolkit 11 lessons + 1 Section Overview: Expanded RAG Toolkit" + }, + { + "span_id": "span-3ce1dddd0035be92640f1a9d", + "text": "2 Non-Technical Skills for Effective LLM Development" + }, + { + "span_id": "span-b62891cbd881c6c9b9e66714", + "text": "3 Adding Question Validation and Routing" + } + ] + }, + { + "chunk_id": "tai-2a1b6a30ca94843de49d55d6", + "text": "Course curriculum\nReasoning Language Models 5 Reinforcement Fine-Tuning for OpenAI 6 Enabling Conversational Memory in LLMs 7 Adding Speech with GPT-4o Audio 8 Long Context LLMs vs. RAG 9 Long Context LLMs & Context Caching vs RAG 10 Graph ML and GraphRAG 11 Quiz 10 Part 1: APIs, UI and Deployment 6 lessons + 1 Section Overview: Building and Deploying the Final RAG Chatbot 2 Bringing it all together: Building the final AI Tutor 3 How to make an API? 4 Building and Deploying a Gradio UI on Hugging Face Spaces 5 Quick Tips for Improving LLM Apps Once in Production 6 Quiz 11 Part 2: More LLM Capabilities and Useful AI Models 6 lessons + 1 Part 2 Intro; Introducing more tools, models & frameworks for LLM projects 2 Section Overview: More LLM Capabilities & Other Useful AI Models 3 Image Generation (Diffusion) and Speech models 4 Generating and Editing Images via the OpenAI API 5 LLM Agents: LLM Pipelines and Tool Use for Complex Tasks 6 Quiz 12 Part 2: More LLM Frameworks and Tools 10 lessons + 1 Section Overview: More LLM Frameworks and Tools 2 More LLM framework options; LangChain vs. LlamaIndex vs. OpenAI Assistants 3 LangChain Introduction 101 4 Access More AI Models with Hugging Face's APIs 5 Building Agents with OpenAI Assistants 6 LangChain OpenGPT 7 Extending LLM Capabilities with Anthropic's Model Context Protocol (MCP) 8 Observability or tracing in production; Langsmith, Phoenix and Langfuse 9 More APIs and tools 10 Quiz 13 Part 2: LLM Optimization 7 lessons + 1 Section Overview: LLM Optimizations for Deployment 2 Best tips for Pruning and Distillation (Minitron Paper) 3 Model Distillation and Teacher-Student Models 4 LLM Deployment Optimisation; Quantization, Pruning, Speculative Decoding 5 Deploying a Quantized LLM on CPU on Google Cloud Platform (GCP) 6", + "heading": "Course curriculum", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-a25942890976c8e8dda3a3ad", + "text": "11 Quiz" + }, + { + "span_id": "span-23557fb9d1ab486f991f64c2", + "text": "5 Reinforcement Fine-Tuning for OpenAI" + }, + { + "span_id": "span-2d191fb52f424b7cf79a4995", + "text": "6 Enabling Conversational Memory in LLMs" + }, + { + "span_id": "span-f7896a588f3263a0e0cfe8ae", + "text": "7 Adding Speech with GPT-4o Audio" + }, + { + "span_id": "span-ed4a96469d4a8c846ace8f3b", + "text": "8 Long Context LLMs vs." + }, + { + "span_id": "span-290dcdde9051137af77d3189", + "text": "9 Long Context LLMs & Context Caching vs RAG" + }, + { + "span_id": "span-437847cdc19ec61ddac53c69", + "text": "10 Graph ML and GraphRAG" + }, + { + "span_id": "span-5567d1134739737822a96cb7", + "text": "10 Part 1: APIs, UI and Deployment 6 lessons + 1 Section Overview: Building and Deploying the Final RAG Chatbot" + }, + { + "span_id": "span-d462797696d44d423ebdd068", + "text": "2 Bringing it all together: Building the final AI Tutor" + }, + { + "span_id": "span-9e459357a328257b219937fb", + "text": "3 How to make an API?" + }, + { + "span_id": "span-8a0ef1bf9b24dec0946347c3", + "text": "4 Building and Deploying a Gradio UI on Hugging Face Spaces" + }, + { + "span_id": "span-47846af735a42358bcb82ccc", + "text": "5 Quick Tips for Improving LLM Apps Once in Production" + }, + { + "span_id": "span-a40381f682ba8c8f4f21ecdf", + "text": "6 Quiz" + }, + { + "span_id": "span-31d2e9010f4da588d5d9d2c4", + "text": "11 Part 2: More LLM Capabilities and Useful AI Models 6 lessons + 1 Part 2 Intro; Introducing more tools, models & frameworks for LLM projects" + }, + { + "span_id": "span-582a71eb59839658480038d4", + "text": "2 Section Overview: More LLM Capabilities & Other Useful AI Models" + }, + { + "span_id": "span-354d3769854b7eae2fe5a96f", + "text": "3 Image Generation (Diffusion) and Speech models" + }, + { + "span_id": "span-a4337f3ddbe0d9ead6e0f5cf", + "text": "4 Generating and Editing Images via the OpenAI API" + }, + { + "span_id": "span-19c08abeb9cead862530c155", + "text": "5 LLM Agents: LLM Pipelines and Tool Use for Complex Tasks" + }, + { + "span_id": "span-fb15c67f80ec4705c4ae9f8f", + "text": "12 Part 2: More LLM Frameworks and Tools 10 lessons + 1 Section Overview: More LLM Frameworks and Tools" + }, + { + "span_id": "span-6bbabbdc181a8108f4798f02", + "text": "2 More LLM framework options; LangChain vs." + }, + { + "span_id": "span-ab0a711aaac2129762180176", + "text": "LlamaIndex vs." + }, + { + "span_id": "span-e753d36ba0211c7fd15bd429", + "text": "OpenAI Assistants" + }, + { + "span_id": "span-81f8bb192e8c1fcb52ac3656", + "text": "3 LangChain Introduction 101" + }, + { + "span_id": "span-972ebf44cde28c7af97ee1a8", + "text": "4 Access More AI Models with Hugging Face's APIs" + }, + { + "span_id": "span-01838c53d7d9e0436cafc240", + "text": "5 Building Agents with OpenAI Assistants" + }, + { + "span_id": "span-be3842ecd16a291ef3073d19", + "text": "6 LangChain OpenGPT" + }, + { + "span_id": "span-618155ed8af7af5099ee8d5a", + "text": "7 Extending LLM Capabilities with Anthropic's Model Context Protocol (MCP)" + }, + { + "span_id": "span-f0f4ec7f4628a665b9168d9a", + "text": "8 Observability or tracing in production; Langsmith, Phoenix and Langfuse" + }, + { + "span_id": "span-c7a9d44b1020c4736cee9129", + "text": "9 More APIs and tools" + }, + { + "span_id": "span-e3e3cf84c7b179021d6a1491", + "text": "10 Quiz" + }, + { + "span_id": "span-9f00e055b11680ea46b83e6b", + "text": "13 Part 2: LLM Optimization 7 lessons + 1 Section Overview: LLM Optimizations for Deployment" + }, + { + "span_id": "span-029f69a7c377bd4d42149c7f", + "text": "2 Best tips for Pruning and Distillation (Minitron Paper)" + }, + { + "span_id": "span-fa923b71abcc804208380a1a", + "text": "3 Model Distillation and Teacher-Student Models" + }, + { + "span_id": "span-e61f1e853023fba91a3f15f3", + "text": "4 LLM Deployment Optimisation; Quantization, Pruning, Speculative Decoding" + }, + { + "span_id": "span-9bbb867eee2c90f6090f33b5", + "text": "5 Deploying a Quantized LLM on CPU on Google Cloud Platform (GCP)" + } + ] + }, + { + "chunk_id": "tai-b08a8fb7a4e0ba4d79d90b4d", + "text": "Course curriculum\nDeploying an LLM Using Latitude.sh 7 Quiz 14 Part 3: LLM Ecosystem Economics; Choosing Your Niche and Business Strategy 10 lessons + 1 Part 3 Intro (optional); Deciding which LLM RAG project to build yourself 2 What will drive Enterprise adoption of Generative AI? 3 Who are the Generative AI Winners? How will Adoption Reshape the Economy? 4 Impact of Large Language Models on Different Job Categories 5 175 LLM Project ideas; Choosing your industry, niche, primary function and target users 6 What Makes a Defensible Product Built with Foundation LLMs? 7 AI LLM/RAG deep dive; Perplexity 8 LLM/RAG deep dive; Consensus 9 AI/LLM Deep dive; Harvey AI 10 Quiz 15 Part 4: Project + Extra Tools 5 lessons + 1 Building Your Own advanced LLM + RAG Project to receive certification 2 Project Submission 3 Share your success and showcase your new skills! 4 AI Tutor: Ask any question! 5 Staying Ahead of AI Developments Updated every two weeks with new lessons as the field evolves. Pay once 3 monthly payments $399 $349 one-time Save 12% Pay once and save $50 versus $399 on the 3-month plan. $133 /month for 3 months $399 in total, with the same lifetime access. Build LLM products end to end: prompting, advanced RAG, fine-tuning, agents Format 92 lessons, 60+ hours, code-heavy and project-based Outcome Certification, a reviewed capstone, and lifetime access with weekly updates Discord community and AI tutor support Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 07 The learning experience", + "heading": "Course curriculum", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-a0bdd5c4a1d4a71b4066e341", + "text": "7 Quiz" + }, + { + "span_id": "span-4ee44ecb9de45f82c295c76d", + "text": "10 Quiz" + }, + { + "span_id": "span-64c331134fef7079a1c33a8d", + "text": "14 Part 3: LLM Ecosystem Economics; Choosing Your Niche and Business Strategy 10 lessons + 1 Part 3 Intro (optional); Deciding which LLM RAG project to build yourself" + }, + { + "span_id": "span-e22ee00874944b4a391835b8", + "text": "2 What will drive Enterprise adoption of Generative AI?" + }, + { + "span_id": "span-fbaeec934188b0ef5e4507e8", + "text": "3 Who are the Generative AI Winners?" + }, + { + "span_id": "span-280c3336f5139707d5c8003a", + "text": "How will Adoption Reshape the Economy?" + }, + { + "span_id": "span-066de577c38c6d337badd80c", + "text": "4 Impact of Large Language Models on Different Job Categories" + }, + { + "span_id": "span-d76a06ff2f3036f3d298ecb2", + "text": "5 175 LLM Project ideas; Choosing your industry, niche, primary function and target users" + }, + { + "span_id": "span-412de63d08b735b4a823c674", + "text": "6 What Makes a Defensible Product Built with Foundation LLMs?" + }, + { + "span_id": "span-94b1de89a6948de96285c048", + "text": "7 AI LLM/RAG deep dive; Perplexity" + }, + { + "span_id": "span-311d6d68ff01c4a94d7ea33f", + "text": "8 LLM/RAG deep dive; Consensus" + }, + { + "span_id": "span-7b1b812f868389bea81cd87b", + "text": "9 AI/LLM Deep dive; Harvey AI" + }, + { + "span_id": "span-81174e75ab06c2e330118694", + "text": "15 Part 4: Project + Extra Tools 5 lessons + 1 Building Your Own advanced LLM + RAG Project to receive certification" + }, + { + "span_id": "span-4f249df657b8830388b655c1", + "text": "2 Project Submission" + }, + { + "span_id": "span-21f814b42d1d679311ad37ee", + "text": "3 Share your success and showcase your new skills!" + }, + { + "span_id": "span-ff31d737be9c5013c0834c3f", + "text": "4 AI Tutor: Ask any question!" + }, + { + "span_id": "span-111c2a27b84615c75ba6a8db", + "text": "5 Staying Ahead of AI Developments" + }, + { + "span_id": "span-5903100db06f93d1525aac61", + "text": "Updated every two weeks with new lessons as the field evolves." + }, + { + "span_id": "span-b6bab6dddb555a2fafbbcaa3", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-3e40df2950caf755e1b7e70f", + "text": "$399 $349 one-time Save 12%" + }, + { + "span_id": "span-2154c68d05d3ca5e516c3986", + "text": "Pay once and save $50 versus $399 on the 3-month plan." + }, + { + "span_id": "span-e9662146d8b3e1b8a566bc34", + "text": "$133 /month for 3 months" + }, + { + "span_id": "span-d389e4b6877d52567e901e46", + "text": "$399 in total, with the same lifetime access." + }, + { + "span_id": "span-639308bdacc3babc278fed08", + "text": "LLM products end to end: prompting, advanced RAG, fine-tuning, agents" + }, + { + "span_id": "span-7785f9f34fa3a68d0b45cc75", + "text": "92 lessons, 60+ hours, code-heavy and project-based" + }, + { + "span_id": "span-3aaa3c707a97d4382406095c", + "text": "Certification, a reviewed capstone, and lifetime access with weekly updates" + }, + { + "span_id": "span-256b734851a462cbbdd7807b", + "text": "Discord community and AI tutor support" + }, + { + "span_id": "span-3f82ef08f56177c96018a4dc", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-c755f7aa0e70156d531d2943", + "text": "07 The learning experience" + } + ] + }, + { + "chunk_id": "tai-36cdce65403047f1cb4a0f20", + "text": "How it works\nProject-based learning designed for working professionals, fully self-paced. 01\n\nSelf-paced, on your schedule\nStart the day you enroll, move at your own pace, and revisit any lesson when a project demands it. 02\n\nLearn by building\nHands-on projects from day one. Build a production LLM system from scratch rather than following along with toy examples. 03\n\nActive Discord community\nGet unstuck fast. Connect with fellow builders, share progress, and collaborate on projects in an active engineering community. 04\n\nCertificate upon completion\nGraduate with certification backed by a real deployed system: portfolio-ready proof you ship. 05\n\nLifetime access and updates\nAll current content plus future updates as AI evolves. Your investment stays valuable. 06\n\nMoney-back guarantee\n30-day full refund, no questions asked. Risk-free enrollment. 08 Before you start\n\nCourse prerequisites\nThe projects are hands-on, so a technical foundation matters. 01 Intermediate Python knowledge 02 Basic knowledge of GitHub 03 A device you can set up a coding environment on 04 At least 50 hours to gain the full benefit of the course New to Python? Start with our Beginner Python for AI Engineering course first. Enrollment\n\nBuild your first scalable product\nWith LLMs, prompting, RAG, tool use, and fine-tuning. $349 one-time, with lifetime access and all future updates. Unlock access for $349 Secure checkout via Towards AI Academy. 30-day money-back guarantee: if you explore the early lessons and it's not what you need, email us for a full refund. 09 Questions", + "heading": "Course prerequisites", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-062e4bdd92ffb379cff14c4b", + "text": "Project-based learning designed for working professionals, fully self-paced." + }, + { + "span_id": "span-f014af5349323fa90e1a79cd", + "text": "Start the day you enroll, move at your own pace, and revisit any lesson when a project demands it." + }, + { + "span_id": "span-d977c958c08be1213d7b0217", + "text": "Hands-on projects from day one." + }, + { + "span_id": "span-9d79786f21342b1869ca25fe", + "text": "Build a production LLM system from scratch rather than following along with toy examples." + }, + { + "span_id": "span-f3d5a27df3408a5daf046857", + "text": "Get unstuck fast." + }, + { + "span_id": "span-7d07c4154a28708bc5c80698", + "text": "Connect with fellow builders, share progress, and collaborate on projects in an active engineering community." + }, + { + "span_id": "span-03632712262cc26f8d4385f6", + "text": "Graduate with certification backed by a real deployed system: portfolio-ready proof you ship." + }, + { + "span_id": "span-75a7496f3090fcc04629cff6", + "text": "All current content plus future updates as AI evolves." + }, + { + "span_id": "span-d583cd4c124e0917fc47a74c", + "text": "Your investment stays valuable." + }, + { + "span_id": "span-ee20dda8b73cb2622575f579", + "text": "30-day full refund, no questions asked." + }, + { + "span_id": "span-d1b37813cbe715b1fa4e630b", + "text": "Risk-free enrollment." + }, + { + "span_id": "span-cabe78eec33e4bd8f25a3f9e", + "text": "08 Before you start" + }, + { + "span_id": "span-0c965fe120e43598a5252c7a", + "text": "The projects are hands-on, so a technical foundation matters." + }, + { + "span_id": "span-7389fe411f7d28862901f8b9", + "text": "01 Intermediate Python knowledge" + }, + { + "span_id": "span-4e285cd383702236efda5b1a", + "text": "02 Basic knowledge of GitHub" + }, + { + "span_id": "span-8551524bd938b8793239db6a", + "text": "03 A device you can set up a coding environment on" + }, + { + "span_id": "span-66854619d9014fe1105662a9", + "text": "04 At least 50 hours to gain the full benefit of the course" + }, + { + "span_id": "span-727573d70a0b9b5419b7d600", + "text": "New to Python?" + }, + { + "span_id": "span-4c2c638e4acb363e0401d058", + "text": "Start with our Beginner Python for AI Engineering course first." + }, + { + "span_id": "span-7487ab76ad1addf22c18a08e", + "text": "With LLMs, prompting, RAG, tool use, and fine-tuning. $349 one-time, with lifetime access and all future updates." + }, + { + "span_id": "span-49de8ef20d81ff312bd98e05", + "text": "Unlock access for $349" + }, + { + "span_id": "span-503f8a5d03de519a3a5a0a9b", + "text": "Secure checkout via Towards AI Academy." + }, + { + "span_id": "span-2d6a7cf053afd523fb14ef7a", + "text": "30-day money-back guarantee: if you explore the early lessons and it's not what you need, email us for a full refund." + }, + { + "span_id": "span-ad44aedb758d48b9b005f120", + "text": "09 Questions" + } + ] + }, + { + "chunk_id": "tai-c07454a29ae38b9aeff34ed3", + "text": "Frequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What projects will I work on? + You'll build your own AI tutor, create RAG pipelines, and ship an advanced LLM MVP. Each project is designed to solve real-world problems, so you graduate with practical, hands-on experience. What are the course prerequisites? + Intermediate Python, basic GitHub knowledge, and a device you can set up a coding environment on. No prior AI or LLM experience needed. When was this course created? Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field. How is this course different from other AI courses? + It is fully hands-on, project-based, and focused on the skills top companies are hiring for right now. We guide you through every step, so you understand the concepts and can apply them immediately. How does this course compare to your \"Agent Engineering\" course? + Full Stack AI Engineering covers the complete LLM product lifecycle: data collection, RAG, fine-tuning, deployment, and business positioning, with agents as one component. The Agents course goes much deeper into designing, orchestrating, evaluating, and deploying autonomous systems. Compare the Agent Engineering course . Do you have a referral or affiliate program? + Yes. You can earn cashback, free courses, and up to $70 per course or $180 per bundle in commission by promoting our courses and the \"Building LLMs for Production\" eBook. See the Affiliate page for tiers and payouts. Can I get a discount? + Yes, there are three ways to receive a discount. Students get 50% off all courses. If you've previously purchased", + "heading": "Frequently asked questions", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-3cf110e6dae929b3bb62a7e8", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-b119a08739045fa14e3b7dae", + "text": "You'll build your own AI tutor, create RAG pipelines, and ship an advanced LLM MVP." + }, + { + "span_id": "span-42cf6be5ddea2cae7eae4cb3", + "text": "Each project is designed to solve real-world problems, so you graduate with practical, hands-on experience." + }, + { + "span_id": "span-f2a709965cde6e2ac450a6df", + "text": "Intermediate Python, basic GitHub knowledge, and a device you can set up a coding environment on." + }, + { + "span_id": "span-47d7fe209a2bdea78c6d1e8c", + "text": "No prior AI or LLM experience needed." + }, + { + "span_id": "span-99c96d56019c2f8a4678375b", + "text": "The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves, so the material keeps pace with the field." + }, + { + "span_id": "span-9c2b773c3af18516aeaff13b", + "text": "It is fully hands-on, project-based, and focused on the skills top companies are hiring for right now." + }, + { + "span_id": "span-9af1b8b0d9cb4bf33e5d9fa8", + "text": "We guide you through every step, so you understand the concepts and can apply them immediately." + }, + { + "span_id": "span-44cd2acaf49d7d26b6777a13", + "text": "You can earn cashback, free courses, and up to $70 per course or $180 per bundle in commission by promoting our courses and the \"Building LLMs for Production\" eBook." + }, + { + "span_id": "span-dee0de681f9bdf085e0ac53e", + "text": "See the Affiliate page for tiers and payouts." + } + ] + }, + { + "chunk_id": "tai-14324bc6318e752f86e50ea1", + "text": "Frequently asked questions\nour book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Can I take this course within my company? + Yes. We offer both course bundles and custom training solutions tailored specifically for companies. For company packages or a customized training plan, reach out to Louis at louis@towardsai.net . Lifetime access · $349\n\nReady to build real products with LLMs?\nJoin 1,000+ AI engineers. Self-paced, updated weekly, and backed by a 30-day money-back guarantee. Unlock lifetime access for $349 Try 6 lessons free", + "heading": "Frequently asked questions", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-12a3f5aa8a7c0930329ff06e", + "text": "We offer both course bundles and custom training solutions tailored specifically for companies." + }, + { + "span_id": "span-57206aa3c7daab4efdc5282d", + "text": "For company packages or a customized training plan, reach out to Louis at louis@towardsai.net ." + }, + { + "span_id": "span-147882e829d4eda02f429a93", + "text": "Lifetime access · $349" + }, + { + "span_id": "span-7a57a60218e6baa1647598ae", + "text": "Join 1,000+ AI engineers." + }, + { + "span_id": "span-faf11943aed3cb32f3a00418", + "text": "Self-paced, updated weekly, and backed by a 30-day money-back guarantee." + }, + { + "span_id": "span-654fcc048905592690dd684a", + "text": "Unlock lifetime access for $349 Try 6 lessons free" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "d7c45e8a3a1216b576823cb356d0080d95ba18454bb91082940a37dd40a1f63c", + "content_hash": "d7c45e8a3a1216b576823cb356d0080d95ba18454bb91082940a37dd40a1f63c", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/full-stack-ai-engineering/" + ], + "evidence_hash": "777ad53e8465fc3967af57fb9351b107a37d560628b13e8b6527df15771975c8" + }, + { + "discovered_url": "https://towardsai.com/academy/agent-engineering/", + "url": "https://towardsai.com/academy/agent-engineering/", + "canonical_url": "https://towardsai.com/academy/agent-engineering/", + "host": "towardsai.com", + "path": "/academy/agent-engineering", + "kind": "course", + "offer_id": "agent-engineering", + "entity_id": "offer:agent-engineering", + "title": "Agent Engineering · Production AI Agents Course | Towards AI Academy", + "review_title": "Agent Engineering", + "reviewed_summary": "Production-focused agent engineering course.", + "meta_description": "Learn agent engineering: design, build, evaluate, and deploy production-grade AI agents and multi-agent systems. 35 lessons, two deployed systems. Lifetime access for $499.", + "headings": [ + "From Prompting Models to Engineering Agents", + "What Students Are Saying About Agent Engineering", + "Where AI Engineers Get Stuck", + "Build two production agents, fully deployed", + "Master the decision framework", + "Build mental models, not framework knowledge", + "From Agent User to Agent Builder", + "Careers Built Through This Course", + "What You'll Build", + "Pure Foundations", + "Four Mini Projects", + "A Deep Research Agent", + "A Deterministic Writing Workflow", + "Multi-Agent Integration", + "Production Stack", + "Capstone Project", + "Learn the Discipline that Turns LLM Calls into Reliable Multi-Agent Systems", + "Course Curriculum", + "Agent Engineering: Building Multi-Agent Systems", + "How It Works", + "Self-Paced + Instructor Support", + "Learn by Building", + "Active Discord Community", + "Certificate Upon Completion", + "Lifetime Access + Updates", + "Money-Back Guarantee", + "Risk-Free Learning: 30-Day Money-Back Guarantee", + "Who This Course Is For", + "Brought to you by", + "Paul Iusztin", + "Build Your First Agent This Week", + "FAQ" + ], + "text": "From Prompting Models to Engineering Agents\nBecome an Agent AI Engineer who designs production-ready systems and leave with an AI agent portfolio that proves it. Try 7 Lessons Free Taught by Louis-François Bouchard & Paul Iusztin Towards AI co-founder & CTO · author, LLM Engineer's Handbook Pay once 3 monthly payments $597 $499 one-time Save 16% Pay once and save $98 versus $597 on the 3-month plan. $199 /month for 3 months $597 in total, with the same lifetime access. Build Two production agents from scratch: reasoning loops, tools, MCP, multi-agent Format 35 lessons with evals, deployment, and CI/CD baked in Outcome Certification backed by a working agent portfolio 5.0 (26 reviews) Unlock Lifetime Access Start the 3-month plan 30-day money-back guarantee, no questions asked\n\nWhat Students Are Saying About Agent Engineering\n\nWhere AI Engineers Get Stuck\nLearn the discipline that turns LLM calls into reliable multi-agent systems. Agents Work Locally, Break in Production\n\nBuild two production agents, fully deployed\nBuild Research Agent and Writing Workflow with full evaluation systems, observability, and deployment pipelines. These are production templates you can adapt for market research, content & report generation, diagrams, documentation, or any domain where autonomous intelligence and automations with agentic patterns add value. Not Sure When Agents Are Worth It\n\nMaster the decision framework\nLearn to design workflows that stay maintainable as requirements evolve. Master when to use deterministic routing versus agent autonomy, when to keep humans in the loop, and the patterns that scale from prototype to production. Skills Get Outdated Fast\n\nBuild mental models, not framework knowledge\nWe teach system design fundamentals from scratch, not just today's frameworks. Graduate with two deployable agents, certification, and the engineering judgment to architect any AI system, regardless of which tools are trending next.\n\nFrom Agent User to Agent Builder\nPure foundations from scratch, 4 mini projects, 2 production systems, deployed end-to-end Start Building\n\nCareers Built Through This Course\n“The curriculum gave me structure, but it also required application. That foundation became part of a larger career transition.” Sean Myers · Course Graduate One graduate's path Before Retail Loss Prevention & Analytics After Sr. Solutions Architect, AI The work behind it What made the difference Repeated exposure to core concepts, practical implementation, structured projects, and a progression from LLM fundamentals into production-grade agent engineering. Certification capstone Knowledge Assistant MCP Server - FastMCP, ChromaDB, RAG, citations, human-in-the-loop, multi-agent coordination. Built entirely in Python\n\nWhat You'll Build\nStart by breaking down complex concepts in interactive notebooks and finish by engineering production-ready code modules that go deeper than frameworks\n\nPure Foundations\nThe mechanics behind agents, workflows, and every tool you'll use in production - taught with code. Architecture & Reasoning Algorithms workflows vs. agents, Chain-of-Thought, ReAct, Plan-and-Execute Context Engineering structured outputs, Pydantic schemas, compression, caching, context selection & isolation Function Calling tool schemas, execution flows, building a @tool decorator framework from scratch RAG, Memory RAG vs. CAG, agentic RAG, multimodal RAG, agent memory with mem0 + ChromaDB building the core patterns from scratch and first principles\n\nFour Mini Projects\nPut every foundation to work - four small builds before production begins: ReAct Practice the observe-think-act loop that powers all modern agents Workflow Engine chaining, routing, parallelization, and orchestrator-worker Multimodal RAG Agent agentic retrieval and reasoning across text, images, and documents based on CLIP & ColPali Harness an agentic harness built as an MCP client that integrates with a deep research agent your first production system\n\nA Deep Research Agent\nBuild an autonomous deep research system to synthesize structured research notes. You'll implement: Multi-Source Data Collection crawl websites, ingest documents, GitHub repositories, and YouTube videos MCP Server Serve the research agent as an MCP server Tool Orchestration integrate Firecrawl, gitingest, Gemini, and Perplexity via function calling Deep Research Algorithm from Scratch parallel multi-round research cycles, automated query generation and refinement, ranking sources Human-in-the-Loop Research approve or redirect the agent's queries at every round your second production system\n\nA Deterministic Writing Workflow\nTransform research into publication-ready articles or lessons. You'll implement: Multi-Media Generation properly formatted text, diagrams, code into a single polished document Prompting for Generating Standardized Outputs context engineering for generating 3000-5000 word content with references Evaluator-Optimizer Pattern automated review and edit loops Orchestrator-Worker Architecture extendable way to generate dozens of media assets in parallel MCP-based Editing Tools real-time human-in-the-loop with agentic tools such as Cursor, Claude, or Codex LangGraph Workflows production pipeline orchestration Structure Projects use a practical version of the clean architecture to structure Python projects easily understood by agentic coding tools such as Claude Code connecting both agents into one multi-agent system\n\nMulti-Agent Integration\nWire the research agent and writing workflow into a unified, orchestrated system. You'll implement: Central Orchestration compose tools from both MCP Servers into a unified agentic system Multi-Server MCP Client one client orchestrating multiple independent MCP servers Composed MCP Server a unified endpoint that proxies both agents via FastMCP Multi-Agent Processes and Loops design multi-agent processes and loops balancing human and agent input ship, monitor, and maintain\n\nProduction Stack\nEverything that separates a demo from production. You'll implement: Observability via Opik tracing, cost, and latency monitoring, debugging non-deterministic agent behavior AI Evals build evaluation datasets, LLM judges, align the judge with the domain expert CI/CD for AI GitHub Actions, mocked LLM tests, AI evals as regression checks, infrastructure as code, automatic deploys to GCP MCP Authentication multi-tenant user support for the MCP server with Descope Deployment Docker, deploying the MCP server to GCP, database integration, autoscaling build your own MCP server\n\nCapstone Project\nShip a portfolio-ready AI project from idea to demo: Choose pick your own idea or start from a curated list Build implement using techniques from the course Iterate get real-time feedback from the instructor as you build Optional Demo present your project on final demo day\n\nLearn the Discipline that Turns LLM Calls into Reliable Multi-Agent Systems\nFrom agent user to agent builder Enroll Now\n\nCourse Curriculum\n01 Part 1: Foundations of Agents and Workflows 17 items + 1 Lesson 0: Build a Research and Writing Agent with MCP Free preview 2 Lesson 1, Part 1: The AI Engineer & The Agent Landscape Free preview 3 Lesson 1, Part 2: Inside the Systems You Build Free preview 4 Lesson 1, Part 3: How to Run the Code Exercises Free preview 5 Lesson 2: LLM Workflows vs. AI Agents -The AI Engineer's Dilemma Free preview 6 Lesson 3: Context Engineering Free preview 7 Lesson 4: Structured Outputs Free preview 8 Quick Quiz 1 Free preview 9 Lesson 5: LLM Workflow Patterns Free preview 10 Lesson 6: Tools Free preview 11 Lesson 7: Planning and Reasoning 12 Quick Quiz 2 13 Lesson 8: ReAct Practice 14 Lesson 9: RAG Focus 15 Lesson 10: Memory for Agents 16 Lesson 11: Multimodal Data 17 Quick Quiz 3 02 Part 2A: Building Agentic Systems; Preparing to Build our Central Research Agent and Writing Workflow 5 items + 1 Lesson 12: Central Project: Scope & Design 2 Lesson 13: Agent Frameworks Overview & Comparison 3 Lesson 14: LLM Agent System Design Considerations and Framework 4 Running the Agents 5 Quick Quiz 4 03 Part 2B: Building Agentic Systems; Building our Central Research Agent 6 items + 1 Lesson 15: Nova End-to-End Project Walkthrough 2 Lesson 16: Foundations of Agentic Systems with FastMCP 3 Lesson 17: Initial Data Ingestion and Tooling 4 Lesson 18: The Research Loop: Query Generation, Perplexity, and Human Feedback 5 Lesson 19: Final Outputs and Agent Completion 6 Quick Quiz 5 04 Part 2C: Building Agentic Systems; Building our Central Writing Workflow 8 items + 1 Lesson 20: Brown End-to-End Project Walkthrough 2 Lesson 21: Behind the Scenes of Iterating AI Architectures with the Brown Writing Agent 3 Lesson 22: Implementing the Foundations of the Writing Workflow 4 Lesson 23: Reviewing and Editing Through the Evaluator-Optimizer Pattern 5 Lesson 24: Human-in-the-Loop Through MCP Servers 6 Lesson 25: Orchestrate and Integrate Our Capstone Agents 7 Quick Quiz 6 8 Lesson 26: End-to-End Demo: Generating a Course Lesson 05 Part 3: Evaluation, Observability, Optimizations, and Deployment 10 items + 1 Lesson 27: Agent Observability with Opik 2 Lesson 28: Creating Datasets for AI Evals 3 Lesson 29: Defining the Evaluation Processes and Metrics Theory 4 Lesson 30: Evaluating the Writing Workflow 5 Quick Quiz 7 6 Lesson 31: Continuous Integration for AI Engineering 7 Lesson 32: Preparing For Deployment: Authentication and Docker 8 Lesson 33: Preparing For Deployment: Database and File Download/Upload 9 Lesson 34: Continuous Deployment for AI Engineering 10 Quick Quiz 8 06 Part 4: Instructions for Students Capstone Project and Certification 3 items + 1 Part 4: Building Your Own MCP Server (To Receive Certification) 2 Part 4: Project Submission 3 Share your success and showcase your new skills! 07 Extra tools 1 item + 1 AI Tutor - Ask any question!\n\nAgent Engineering: Building Multi-Agent Systems\nPay once 3 monthly payments $597 $499 one-time Save 16% Pay once and save $98 versus $597 on the 3-month plan. $199 /month for 3 months $597 in total, with the same lifetime access. Build Two production agents from scratch: reasoning loops, tools, MCP, multi-agent Format 35 lessons with evals, deployment, and CI/CD baked in Outcome Certification backed by a working agent portfolio 5.0 (26 reviews) Unlock Lifetime Access Start the 3-month plan Try 7 Lessons Free\n\nHow It Works\nSelf-paced project-based learning designed for working professionals. 01\n\nSelf-Paced + Instructor Support\nLearn on your schedule with lifetime access. Join dedicated Discord channels with instructors and fellow course takers. 02\n\nLearn by Building\nHands-on projects from day one. Build two production agents from scratch, not toy examples or follow-along tutorials. 03\n\nActive Discord Community\nGet unstuck fast. Connect with fellow builders, share progress, and collaborate on projects in an active engineering community. 04\n\nCertificate Upon Completion\nProve you built and deployed production AI. Graduate with certification backed by two real systems and portfolio-ready proof you ship. 05\n\nLifetime Access + Updates\nAll current content plus future updates as AI evolves. Your investment stays valuable. 06\n\nMoney-Back Guarantee\n30-day full refund, no questions asked. Risk-free enrollment.\n\nRisk-Free Learning: 30-Day Money-Back Guarantee\nIf you explore the early part of the course and feel it’s not what you need, you can request a refund within 30 days by emailing us at louis@towardsai.net Start Risk-Free\n\nWho This Course Is For\nFor software engineers, data engineers, data scientists transitioning into AI Engineering 01 Python: Comfortable with functions, classes, and API calls 02 LLMs: Familiar with OpenAI/Claude APIs and basic prompting 03 Docker & Cloud: Understand containerization and deployment basics 04 Skill Level: Intermediate to advanced, expect a challenge 05 Mindset: Learn by building, not watching tutorials\n\nBrought to you by\nTowards AI Towards AI's mission is to make \"building with AI\" more accessible. Since 2019, we have helped teach over 500k about AI with our courses, blogs, tutorials, books, community, and personalized consultancy. Co-instructor\n\nPaul Iusztin\nAfter shipping 21 AI apps, Paul Iusztin has used his best-selling LLM Engineer's Handbook, Decoding AI Magazine, and social media platforms to lead 162,000+ AI engineers out of \"demo purgatory\" and into production-grade engineering.\n\nBuild Your First Agent This Week\nStop calling LLMs. Start engineering agents. Own two agents, fully production-ready Gain portfolio-ready skills with certification Master fundamentals that outlast trends Enroll Now Self-paced learning · Regular updates · Risk-free enrollment\n\nFAQ\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. How does it compare to your \"Full Stack AI Engineering\" course? + The Full Stack AI Engineering course teaches the complete LLM product stack including RAG, fine-tuning, deployment, and agents through one central AI product. The Agents course goes much deeper into designing, orchestrating, evaluating, and deploying production-grade autonomous systems. If Full Stack teaches you how to build an AI product, Agents teaches you how to engineer the autonomy inside it. Compare the Full Stack course . What projects will I work on? + You'll build two interconnected production agents from the ground up: a Research Agent and a Writing Workflow, with human-in-the-loop capabilities, while mastering the foundations of agentic architecture. How is this course different from other AI courses? + Unlike many theoretical courses, this one is fully hands-on, project-based, and focused on the skills top companies are hiring for right now. We guide you through every step, ensuring you not only understand the concepts but can also apply them immediately. Do you have a referral or affiliate program? + For every referral you provide, you’ll gain access to our exclusive eBook. If you refer three or more people, we’ll send you a physical copy of our book as a thank you! Additionally, we have an affiliate program for individuals with an audience. By joining, you can earn commissions for every successful referral made through your unique affiliate link. Please email Louis-François at louis@towardsai.net with proof of referral. Can I get a discount? + Yes, there are three ways to receive a discount on our courses. If you're a student, we offer a 50% discount on all courses. Additionally, if you've previously purchased our book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more people. For further details or to claim your discount, please email Louis-François at louis@towardsai.net for a follow-up. Can I take this course within my company? + Yes! We offer both course bundles and custom training solutions tailored specifically for companies. For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net . Our refund policy + If you feel the course isn’t right for you, we’ll provide a 100% refund within 30 days—no hassle, no worries.", + "links": [ + { + "text": "Try 7 Lessons Free", + "url": "https://towardsai.com/academy/agent-engineering-free-preview/" + }, + { + "text": "Unlock Lifetime Access", + "url": "https://academy.towardsai.net/cart/add_product/3406924?price_id=4319175" + }, + { + "text": "Start the 3-month plan", + "url": "https://academy.towardsai.net/enroll/3406924?price_id=4813876" + }, + { + "text": "Start Building", + "url": "https://academy.towardsai.net/cart/add_product/3406924?price_id=4319175" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-foundations.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-mini-projects.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-research-agent.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-writing-workflow.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-multi-agent.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-production.png" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/agents-build-capstone.png" + }, + { + "text": "Enroll Now", + "url": "https://academy.towardsai.net/cart/add_product/3406924?price_id=4319175" + }, + { + "text": "Lesson 0: Build a Research and Writing Agent with MCP Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 1, Part 1: The AI Engineer & The Agent Landscape Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 1, Part 2: Inside the Systems You Build Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 1, Part 3: How to Run the Code Exercises Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 2: LLM Workflows vs. AI Agents -The AI Engineer's Dilemma Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 3: Context Engineering Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 4: Structured Outputs Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Quick Quiz 1 Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 5: LLM Workflow Patterns Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Lesson 6: Tools Free preview", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + }, + { + "text": "Start Risk-Free", + "url": "https://academy.towardsai.net/cart/add_product/3406924?price_id=4319175" + }, + { + "text": "", + "url": "https://towardsai.com/academy-assets/brown-nova-demo.gif" + }, + { + "text": "Compare the Full Stack course", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + } + ], + "chunks": [ + { + "chunk_id": "tai-bf95fd1ec627c842bcb71e6a", + "text": "From Prompting Models to Engineering Agents\nBecome an Agent AI Engineer who designs production-ready systems and leave with an AI agent portfolio that proves it. Try 7 Lessons Free Taught by Louis-François Bouchard & Paul Iusztin Towards AI co-founder & CTO · author, LLM Engineer's Handbook Pay once 3 monthly payments $597 $499 one-time Save 16% Pay once and save $98 versus $597 on the 3-month plan. $199 /month for 3 months $597 in total, with the same lifetime access. Build Two production agents from scratch: reasoning loops, tools, MCP, multi-agent Format 35 lessons with evals, deployment, and CI/CD baked in Outcome Certification backed by a working agent portfolio 5.0 (26 reviews) Unlock Lifetime Access Start the 3-month plan 30-day money-back guarantee, no questions asked\n\nWhat Students Are Saying About Agent Engineering\n\nWhere AI Engineers Get Stuck\nLearn the discipline that turns LLM calls into reliable multi-agent systems. Agents Work Locally, Break in Production\n\nBuild two production agents, fully deployed\nBuild Research Agent and Writing Workflow with full evaluation systems, observability, and deployment pipelines. These are production templates you can adapt for market research, content & report generation, diagrams, documentation, or any domain where autonomous intelligence and automations with agentic patterns add value. Not Sure When Agents Are Worth It\n\nMaster the decision framework\nLearn to design workflows that stay maintainable as requirements evolve. Master when to use deterministic routing versus agent autonomy, when to keep humans in the loop, and the patterns that scale from prototype to production. Skills Get Outdated Fast", + "heading": "From Prompting Models to Engineering Agents", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-9fc27c44cee1aa0a213bece7", + "text": "Become an Agent AI Engineer who designs production-ready systems and leave with an AI agent portfolio that proves it." + }, + { + "span_id": "span-c6ba2ee33a3f32522304a688", + "text": "Try 7 Lessons Free" + }, + { + "span_id": "span-a0eb49eebc5b3ae5a198fc89", + "text": "Taught by Louis-François Bouchard & Paul Iusztin Towards AI co-founder & CTO · author, LLM Engineer's Handbook" + }, + { + "span_id": "span-7d87fce2ab99b1f183bbbac8", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-27f762aa25a5e60a5d671b0f", + "text": "$597 $499 one-time Save 16%" + }, + { + "span_id": "span-245cb94398d1eebb9e0622da", + "text": "Pay once and save $98 versus $597 on the 3-month plan." + }, + { + "span_id": "span-d5468a776e2027d9cd02bfd8", + "text": "$199 /month for 3 months" + }, + { + "span_id": "span-e15f309d3e073a3cc75849ad", + "text": "$597 in total, with the same lifetime access." + }, + { + "span_id": "span-a11628a5d9eda5e3cbfe24b5", + "text": "Two production agents from scratch: reasoning loops, tools, MCP, multi-agent" + }, + { + "span_id": "span-42e8098b99ba56d561a7240c", + "text": "35 lessons with evals, deployment, and CI/CD baked in" + }, + { + "span_id": "span-e6e61f7cbfc95014c96fedd9", + "text": "Certification backed by a working agent portfolio" + }, + { + "span_id": "span-9c23be16624c8f6c0fab357e", + "text": "5.0 (26 reviews)" + }, + { + "span_id": "span-73d53d7c77c81c2d14e2c489", + "text": "Unlock Lifetime Access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-2145020c9d23486bea603855", + "text": "What Students Are Saying About Agent Engineering" + }, + { + "span_id": "span-af6a8637a98fc08f675d267d", + "text": "Learn the discipline that turns LLM calls into reliable multi-agent systems." + }, + { + "span_id": "span-1b4b0bde596a747ba1e501b5", + "text": "Agents Work Locally, Break in Production" + }, + { + "span_id": "span-e04f17145e36f2c286a75c1e", + "text": "Build Research Agent and Writing Workflow with full evaluation systems, observability, and deployment pipelines." + }, + { + "span_id": "span-66ff08d1836143f2b4089da8", + "text": "These are production templates you can adapt for market research, content & report generation, diagrams, documentation, or any domain where autonomous intelligence and automations with agentic patterns add value." + }, + { + "span_id": "span-ebc3e8c6c3142d93749dabf4", + "text": "Not Sure When Agents Are Worth It" + }, + { + "span_id": "span-cb3b791e52f6086b547b12af", + "text": "Learn to design workflows that stay maintainable as requirements evolve." + }, + { + "span_id": "span-c30527a2167336797c9af28b", + "text": "Master when to use deterministic routing versus agent autonomy, when to keep humans in the loop, and the patterns that scale from prototype to production." + }, + { + "span_id": "span-d0b53dda2073880ab60e9dcf", + "text": "Skills Get Outdated Fast" + } + ] + }, + { + "chunk_id": "tai-7818d7ca941424105404ffb9", + "text": "Build mental models, not framework knowledge\nWe teach system design fundamentals from scratch, not just today's frameworks. Graduate with two deployable agents, certification, and the engineering judgment to architect any AI system, regardless of which tools are trending next.\n\nFrom Agent User to Agent Builder\nPure foundations from scratch, 4 mini projects, 2 production systems, deployed end-to-end Start Building\n\nCareers Built Through This Course\n“The curriculum gave me structure, but it also required application. That foundation became part of a larger career transition.” Sean Myers · Course Graduate One graduate's path Before Retail Loss Prevention & Analytics After Sr. Solutions Architect, AI The work behind it What made the difference Repeated exposure to core concepts, practical implementation, structured projects, and a progression from LLM fundamentals into production-grade agent engineering. Certification capstone Knowledge Assistant MCP Server - FastMCP, ChromaDB, RAG, citations, human-in-the-loop, multi-agent coordination. Built entirely in Python\n\nWhat You'll Build\nStart by breaking down complex concepts in interactive notebooks and finish by engineering production-ready code modules that go deeper than frameworks", + "heading": "Careers Built Through This Course", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-1a805eaf01ea1e4ee32879a7", + "text": "We teach system design fundamentals from scratch, not just today's frameworks." + }, + { + "span_id": "span-91d3310d9ed69aee1e1dc7a4", + "text": "Graduate with two deployable agents, certification, and the engineering judgment to architect any AI system, regardless of which tools are trending next." + }, + { + "span_id": "span-513f3489fd3d2833849dca37", + "text": "Pure foundations from scratch, 4 mini projects, 2 production systems, deployed end-to-end" + }, + { + "span_id": "span-f2bc93aa95979d50d30ca3fe", + "text": "Start Building" + }, + { + "span_id": "span-f3a35ccb2c93e8244a6229d5", + "text": "“The curriculum gave me structure, but it also required application." + }, + { + "span_id": "span-88a668345100d31485870f2a", + "text": "That foundation became part of a larger career transition.”" + }, + { + "span_id": "span-34123f3aa11ef0f7d7f4c966", + "text": "Sean Myers · Course Graduate" + }, + { + "span_id": "span-bc4664bcceea74afa88c2c3b", + "text": "One graduate's path" + }, + { + "span_id": "span-b19adebd4e85e9d990c32e7b", + "text": "Retail Loss Prevention & Analytics" + }, + { + "span_id": "span-fe33d7416b2be4cdc473db24", + "text": "Solutions Architect, AI" + }, + { + "span_id": "span-b0b24a1f189b3d78b22b70f6", + "text": "The work behind it" + }, + { + "span_id": "span-096e391e8e3609271e552b79", + "text": "What made the difference" + }, + { + "span_id": "span-181ce2a1665724565c128164", + "text": "Repeated exposure to core concepts, practical implementation, structured projects, and a progression from LLM fundamentals into production-grade agent engineering." + }, + { + "span_id": "span-9c64abbabe9109f3fa03d393", + "text": "Certification capstone" + }, + { + "span_id": "span-8ef85382f4f0c31cffcbac57", + "text": "Knowledge Assistant MCP Server - FastMCP, ChromaDB, RAG, citations, human-in-the-loop, multi-agent coordination." + }, + { + "span_id": "span-06d8d225b4bfad996b51808f", + "text": "Built entirely in Python" + }, + { + "span_id": "span-57ccc1a54682efbc03e3190a", + "text": "Start by breaking down complex concepts in interactive notebooks and finish by engineering production-ready code modules" + }, + { + "span_id": "span-2bbda9061c58339afab22076", + "text": "that go deeper than frameworks" + } + ] + }, + { + "chunk_id": "tai-da344756f7c379209811d735", + "text": "Pure Foundations\nThe mechanics behind agents, workflows, and every tool you'll use in production - taught with code. Architecture & Reasoning Algorithms workflows vs. agents, Chain-of-Thought, ReAct, Plan-and-Execute Context Engineering structured outputs, Pydantic schemas, compression, caching, context selection & isolation Function Calling tool schemas, execution flows, building a @tool decorator framework from scratch RAG, Memory RAG vs. CAG, agentic RAG, multimodal RAG, agent memory with mem0 + ChromaDB building the core patterns from scratch and first principles\n\nFour Mini Projects\nPut every foundation to work - four small builds before production begins: ReAct Practice the observe-think-act loop that powers all modern agents Workflow Engine chaining, routing, parallelization, and orchestrator-worker Multimodal RAG Agent agentic retrieval and reasoning across text, images, and documents based on CLIP & ColPali Harness an agentic harness built as an MCP client that integrates with a deep research agent your first production system\n\nA Deep Research Agent\nBuild an autonomous deep research system to synthesize structured research notes. You'll implement: Multi-Source Data Collection crawl websites, ingest documents, GitHub repositories, and YouTube videos MCP Server Serve the research agent as an MCP server Tool Orchestration integrate Firecrawl, gitingest, Gemini, and Perplexity via function calling Deep Research Algorithm from Scratch parallel multi-round research cycles, automated query generation and refinement, ranking sources Human-in-the-Loop Research approve or redirect the agent's queries at every round your second production system", + "heading": "A Deep Research Agent", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-b6ef7e574ab674b11ec44ac1", + "text": "The mechanics behind agents, workflows, and every tool you'll use in production - taught with code." + }, + { + "span_id": "span-6c0338b178ce83bc7c5feee1", + "text": "Architecture & Reasoning Algorithms workflows vs. agents, Chain-of-Thought, ReAct, Plan-and-Execute" + }, + { + "span_id": "span-e05f30e84c7797dbbca94af5", + "text": "Context Engineering structured outputs, Pydantic schemas, compression, caching, context selection & isolation" + }, + { + "span_id": "span-0be5c4e25b82b6c920a4474a", + "text": "Function Calling tool schemas, execution flows, building a @tool decorator framework from scratch" + }, + { + "span_id": "span-3d7b8661cdf09314f7dda08c", + "text": "RAG, Memory RAG vs." + }, + { + "span_id": "span-5e3f5b21ec93696cf50b2bcb", + "text": "CAG, agentic RAG, multimodal RAG, agent memory with mem0 + ChromaDB" + }, + { + "span_id": "span-4302c865e4c16d915799a561", + "text": "building the core patterns from scratch and first principles" + }, + { + "span_id": "span-de49bd079622dec13e329991", + "text": "Put every foundation to work - four small builds before production begins:" + }, + { + "span_id": "span-748a8eadc44b9e587511d48c", + "text": "ReAct Practice the observe-think-act loop that powers all modern agents" + }, + { + "span_id": "span-a859cbe728d2353ff1aecf57", + "text": "Workflow Engine chaining, routing, parallelization, and orchestrator-worker" + }, + { + "span_id": "span-d4961bfd9eaac8bb79702ee8", + "text": "Multimodal RAG Agent agentic retrieval and reasoning across text, images, and documents based on CLIP & ColPali" + }, + { + "span_id": "span-4427907ee4ba3ef983ea5241", + "text": "Harness an agentic harness built as an MCP client that integrates with a deep research agent" + }, + { + "span_id": "span-35bbf3d886ace79e3ce0c551", + "text": "your first production system" + }, + { + "span_id": "span-a7d351b7feb1bd3b31e8f1c0", + "text": "Build an autonomous deep research system to synthesize structured research notes." + }, + { + "span_id": "span-12430c8cd0b755f1a1f65173", + "text": "You'll implement:" + }, + { + "span_id": "span-20a3450e7288da5e1572b12b", + "text": "Multi-Source Data Collection crawl websites, ingest documents, GitHub repositories, and YouTube videos" + }, + { + "span_id": "span-0ea96ad1b0a4266006af22ba", + "text": "MCP Server Serve the research agent as an MCP server" + }, + { + "span_id": "span-02bb9858ad2d68cbeb179d2f", + "text": "Tool Orchestration integrate Firecrawl, gitingest, Gemini, and Perplexity via function calling" + }, + { + "span_id": "span-3d87b7b1996fd4f79cb33da6", + "text": "Deep Research Algorithm from Scratch parallel multi-round research cycles, automated query generation and refinement, ranking sources" + }, + { + "span_id": "span-d3555aff89bf66eb1378e7c8", + "text": "Human-in-the-Loop Research approve or redirect the agent's queries at every round" + }, + { + "span_id": "span-ecb001a7af85359d58970119", + "text": "your second production system" + } + ] + }, + { + "chunk_id": "tai-1e7fad0f2b6afc6a165a86e8", + "text": "A Deterministic Writing Workflow\nTransform research into publication-ready articles or lessons. You'll implement: Multi-Media Generation properly formatted text, diagrams, code into a single polished document Prompting for Generating Standardized Outputs context engineering for generating 3000-5000 word content with references Evaluator-Optimizer Pattern automated review and edit loops Orchestrator-Worker Architecture extendable way to generate dozens of media assets in parallel MCP-based Editing Tools real-time human-in-the-loop with agentic tools such as Cursor, Claude, or Codex LangGraph Workflows production pipeline orchestration Structure Projects use a practical version of the clean architecture to structure Python projects easily understood by agentic coding tools such as Claude Code connecting both agents into one multi-agent system\n\nMulti-Agent Integration\nWire the research agent and writing workflow into a unified, orchestrated system. You'll implement: Central Orchestration compose tools from both MCP Servers into a unified agentic system Multi-Server MCP Client one client orchestrating multiple independent MCP servers Composed MCP Server a unified endpoint that proxies both agents via FastMCP Multi-Agent Processes and Loops design multi-agent processes and loops balancing human and agent input ship, monitor, and maintain", + "heading": "A Deterministic Writing Workflow", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-1c83417e2d3aa4cc2f01d0c9", + "text": "Transform research into publication-ready articles or lessons." + }, + { + "span_id": "span-863aa2650504319fcdad664d", + "text": "You'll implement:" + }, + { + "span_id": "span-1145512f60d77172e3bbe2b3", + "text": "Multi-Media Generation properly formatted text, diagrams, code into a single polished document" + }, + { + "span_id": "span-5fce79bea22bd46fe21c48dd", + "text": "Prompting for Generating Standardized Outputs context engineering for generating 3000-5000 word content with references" + }, + { + "span_id": "span-9f523d7ea14a37522fbf0d97", + "text": "Evaluator-Optimizer Pattern automated review and edit loops" + }, + { + "span_id": "span-1cbe7075b0ef40f64b36906f", + "text": "Orchestrator-Worker Architecture extendable way to generate dozens of media assets in parallel" + }, + { + "span_id": "span-e56182c0a86b4b13995c75a3", + "text": "MCP-based Editing Tools real-time human-in-the-loop with agentic tools such as Cursor, Claude, or Codex" + }, + { + "span_id": "span-424ee9fb72c29bdfd3f3771f", + "text": "LangGraph Workflows production pipeline orchestration" + }, + { + "span_id": "span-8ec4729a69359f45fb9532a3", + "text": "Structure Projects use a practical version of the clean architecture to structure Python projects easily understood by agentic coding tools such as Claude Code" + }, + { + "span_id": "span-681f107209bb2cb95f963963", + "text": "connecting both agents into one multi-agent system" + }, + { + "span_id": "span-11e699fd43b4da973dc629e1", + "text": "Wire the research agent and writing workflow into a unified, orchestrated system." + }, + { + "span_id": "span-f0e017089cb2c6399a14ebb3", + "text": "Central Orchestration compose tools from both MCP Servers into a unified agentic system" + }, + { + "span_id": "span-e82db254a16deb320fc3b629", + "text": "Multi-Server MCP Client one client orchestrating multiple independent MCP servers" + }, + { + "span_id": "span-718e380b0633da2b4eaf27e7", + "text": "Composed MCP Server a unified endpoint that proxies both agents via FastMCP" + }, + { + "span_id": "span-05dbe4ec48df959b8d51f325", + "text": "Multi-Agent Processes and Loops design multi-agent processes and loops balancing human and agent input" + }, + { + "span_id": "span-8777952d2f7e176f53482aa9", + "text": "ship, monitor, and maintain" + } + ] + }, + { + "chunk_id": "tai-5bbf1a5c92b65b7c03142d49", + "text": "Production Stack\nEverything that separates a demo from production. You'll implement: Observability via Opik tracing, cost, and latency monitoring, debugging non-deterministic agent behavior AI Evals build evaluation datasets, LLM judges, align the judge with the domain expert CI/CD for AI GitHub Actions, mocked LLM tests, AI evals as regression checks, infrastructure as code, automatic deploys to GCP MCP Authentication multi-tenant user support for the MCP server with Descope Deployment Docker, deploying the MCP server to GCP, database integration, autoscaling build your own MCP server\n\nCapstone Project\nShip a portfolio-ready AI project from idea to demo: Choose pick your own idea or start from a curated list Build implement using techniques from the course Iterate get real-time feedback from the instructor as you build Optional Demo present your project on final demo day\n\nLearn the Discipline that Turns LLM Calls into Reliable Multi-Agent Systems\nFrom agent user to agent builder Enroll Now", + "heading": "Production Stack", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-1e8c61eaed1652fac217b1d2", + "text": "Everything that separates a demo from production." + }, + { + "span_id": "span-378d4ed1b6f19939013b2de9", + "text": "You'll implement:" + }, + { + "span_id": "span-cd5ee7c24d8d03ae69a3cb54", + "text": "Observability via Opik tracing, cost, and latency monitoring, debugging non-deterministic agent behavior" + }, + { + "span_id": "span-350b23d27db1c75c6f8e273a", + "text": "AI Evals build evaluation datasets, LLM judges, align the judge with the domain expert" + }, + { + "span_id": "span-5d743fa7dd1113062e7e71c4", + "text": "CI/CD for AI GitHub Actions, mocked LLM tests, AI evals as regression checks, infrastructure as code, automatic deploys to GCP" + }, + { + "span_id": "span-510422df2ff65eb54566f3cc", + "text": "MCP Authentication multi-tenant user support for the MCP server with Descope" + }, + { + "span_id": "span-c4b6189a81ee32b0c7791408", + "text": "Deployment Docker, deploying the MCP server to GCP, database integration, autoscaling" + }, + { + "span_id": "span-d9405657c70107e8c5be8999", + "text": "build your own MCP server" + }, + { + "span_id": "span-9c1212f20f57ebf274eaadba", + "text": "Ship a portfolio-ready AI project from idea to demo:" + }, + { + "span_id": "span-78ab273b11bb72ea4a51fc0e", + "text": "Choose pick your own idea or start from a curated list" + }, + { + "span_id": "span-f6700d9b552c142f68dfb083", + "text": "Build implement using techniques from the course" + }, + { + "span_id": "span-cc474ebdb2dc7ea512faeaeb", + "text": "Iterate get real-time feedback from the instructor as you build" + }, + { + "span_id": "span-c4630a493c03680b9f36ed83", + "text": "Optional Demo present your project on final demo day" + }, + { + "span_id": "span-c555fdc2ac0e667d8fb81a03", + "text": "From agent user to agent builder" + }, + { + "span_id": "span-da20933ba8dbc7d5340e176e", + "text": "Enroll Now" + } + ] + }, + { + "chunk_id": "tai-8dd6ae8ea8e225abde96ba5e", + "text": "Course Curriculum\n01 Part 1: Foundations of Agents and Workflows 17 items + 1 Lesson 0: Build a Research and Writing Agent with MCP Free preview 2 Lesson 1, Part 1: The AI Engineer & The Agent Landscape Free preview 3 Lesson 1, Part 2: Inside the Systems You Build Free preview 4 Lesson 1, Part 3: How to Run the Code Exercises Free preview 5 Lesson 2: LLM Workflows vs. AI Agents -The AI Engineer's Dilemma Free preview 6 Lesson 3: Context Engineering Free preview 7 Lesson 4: Structured Outputs Free preview 8 Quick Quiz 1 Free preview 9 Lesson 5: LLM Workflow Patterns Free preview 10 Lesson 6: Tools Free preview 11 Lesson 7: Planning and Reasoning 12 Quick Quiz 2 13 Lesson 8: ReAct Practice 14 Lesson 9: RAG Focus 15 Lesson 10: Memory for Agents 16 Lesson 11: Multimodal Data 17 Quick Quiz 3 02 Part 2A: Building Agentic Systems; Preparing to Build our Central Research Agent and Writing Workflow 5 items + 1 Lesson 12: Central Project: Scope & Design 2 Lesson 13: Agent Frameworks Overview & Comparison 3 Lesson 14: LLM Agent System Design Considerations and Framework 4 Running the Agents 5 Quick Quiz 4 03 Part 2B: Building Agentic Systems; Building our Central Research Agent 6 items + 1 Lesson 15: Nova End-to-End Project Walkthrough 2 Lesson 16: Foundations of Agentic Systems with FastMCP 3 Lesson 17: Initial Data Ingestion and Tooling 4 Lesson 18: The Research Loop: Query Generation, Perplexity, and Human Feedback 5 Lesson 19: Final Outputs and Agent Completion 6 Quick Quiz 5 04 Part 2C: Building Agentic Systems; Building our Central Writing Workflow 8 items + 1 Lesson 20: Brown End-to-End Project Walkthrough 2 Lesson 21: Behind the Scenes of Iterating AI Architectures with the Brown Writing Agent 3 Lesson 22: Implementing the Foundations of the Writing Workflow 4 Lesson 23:", + "heading": "Course Curriculum", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-df2a98d39a484135d2c6c12f", + "text": "01 Part 1: Foundations of Agents and Workflows 17 items + 1 Lesson 0: Build a Research and Writing Agent with MCP Free preview" + }, + { + "span_id": "span-565c27c4d58a6678fcea17ae", + "text": "2 Lesson 1, Part 1: The AI Engineer & The Agent Landscape Free preview" + }, + { + "span_id": "span-a88aa163f33c7004ad02f2a3", + "text": "3 Lesson 1, Part 2: Inside the Systems You Build Free preview" + }, + { + "span_id": "span-19fdf8ac851605bd62d94065", + "text": "4 Lesson 1, Part 3: How to Run the Code Exercises Free preview" + }, + { + "span_id": "span-b199adc10634468a70599982", + "text": "5 Lesson 2: LLM Workflows vs." + }, + { + "span_id": "span-169fc224b852e5d4a711cb6a", + "text": "AI Agents -The AI Engineer's Dilemma Free preview" + }, + { + "span_id": "span-e24fd988315836b03c4c6d73", + "text": "6 Lesson 3: Context Engineering Free preview" + }, + { + "span_id": "span-6722ad3123b65551690e5811", + "text": "7 Lesson 4: Structured Outputs Free preview" + }, + { + "span_id": "span-a507af0e843d0e65120d5562", + "text": "8 Quick Quiz 1 Free preview" + }, + { + "span_id": "span-280fea54e550b0341069e8da", + "text": "9 Lesson 5: LLM Workflow Patterns Free preview" + }, + { + "span_id": "span-906651233cec056d8ca8ee79", + "text": "10 Lesson 6: Tools Free preview" + }, + { + "span_id": "span-6923acda2e87ad55e01074e0", + "text": "11 Lesson 7: Planning and Reasoning" + }, + { + "span_id": "span-9e036f635d0332b66ec20b80", + "text": "12 Quick Quiz 2" + }, + { + "span_id": "span-29a21002a76ea24100543cfd", + "text": "13 Lesson 8: ReAct Practice" + }, + { + "span_id": "span-0e206669f5463eb351adaaea", + "text": "14 Lesson 9: RAG Focus" + }, + { + "span_id": "span-7a1ab573d7d427ad60d30e79", + "text": "15 Lesson 10: Memory for Agents" + }, + { + "span_id": "span-d72421cf8bbbec3b34804672", + "text": "16 Lesson 11: Multimodal Data" + }, + { + "span_id": "span-4ae44c482efff5d7e6b51316", + "text": "17 Quick Quiz 3" + }, + { + "span_id": "span-f1447772f2ec8683d6b61fab", + "text": "02 Part 2A: Building Agentic Systems; Preparing to Build our Central Research Agent and Writing Workflow 5 items + 1 Lesson 12: Central Project: Scope & Design" + }, + { + "span_id": "span-adea0b7efa2288740c2a776e", + "text": "2 Lesson 13: Agent Frameworks Overview & Comparison" + }, + { + "span_id": "span-f5a8a2eb6c9b390407f2d313", + "text": "3 Lesson 14: LLM Agent System Design Considerations and Framework" + }, + { + "span_id": "span-ae1cbc1d4c8c2a8a3de8b067", + "text": "4 Running the Agents" + }, + { + "span_id": "span-7fc8804fba11595619489c15", + "text": "5 Quick Quiz 4" + }, + { + "span_id": "span-46562d150d1c2e1fc4f02341", + "text": "03 Part 2B: Building Agentic Systems; Building our Central Research Agent 6 items + 1 Lesson 15: Nova End-to-End Project Walkthrough" + }, + { + "span_id": "span-6a89f3df07b460c61fa996a5", + "text": "2 Lesson 16: Foundations of Agentic Systems with FastMCP" + }, + { + "span_id": "span-94c843908483bd6402fbd633", + "text": "3 Lesson 17: Initial Data Ingestion and Tooling" + }, + { + "span_id": "span-1bff92dd8ba18d9289d467c6", + "text": "4 Lesson 18: The Research Loop: Query Generation, Perplexity, and Human Feedback" + }, + { + "span_id": "span-2a57264beaf5c40d746fb623", + "text": "5 Lesson 19: Final Outputs and Agent Completion" + }, + { + "span_id": "span-dc9d48d9c0bd61a97a8d3e72", + "text": "6 Quick Quiz 5" + }, + { + "span_id": "span-5b1ef9c3110e6404cc0d0e2b", + "text": "04 Part 2C: Building Agentic Systems; Building our Central Writing Workflow 8 items + 1 Lesson 20: Brown End-to-End Project Walkthrough" + }, + { + "span_id": "span-9b6392e40ce7da50619be519", + "text": "2 Lesson 21: Behind the Scenes of Iterating AI Architectures with the Brown Writing Agent" + }, + { + "span_id": "span-eebb1bafd40348e386819024", + "text": "3 Lesson 22: Implementing the Foundations of the Writing Workflow" + } + ] + }, + { + "chunk_id": "tai-0293234be68d1fdc741f771c", + "text": "Course Curriculum\nReviewing and Editing Through the Evaluator-Optimizer Pattern 5 Lesson 24: Human-in-the-Loop Through MCP Servers 6 Lesson 25: Orchestrate and Integrate Our Capstone Agents 7 Quick Quiz 6 8 Lesson 26: End-to-End Demo: Generating a Course Lesson 05 Part 3: Evaluation, Observability, Optimizations, and Deployment 10 items + 1 Lesson 27: Agent Observability with Opik 2 Lesson 28: Creating Datasets for AI Evals 3 Lesson 29: Defining the Evaluation Processes and Metrics Theory 4 Lesson 30: Evaluating the Writing Workflow 5 Quick Quiz 7 6 Lesson 31: Continuous Integration for AI Engineering 7 Lesson 32: Preparing For Deployment: Authentication and Docker 8 Lesson 33: Preparing For Deployment: Database and File Download/Upload 9 Lesson 34: Continuous Deployment for AI Engineering 10 Quick Quiz 8 06 Part 4: Instructions for Students Capstone Project and Certification 3 items + 1 Part 4: Building Your Own MCP Server (To Receive Certification) 2 Part 4: Project Submission 3 Share your success and showcase your new skills! 07 Extra tools 1 item + 1 AI Tutor - Ask any question!\n\nAgent Engineering: Building Multi-Agent Systems\nPay once 3 monthly payments $597 $499 one-time Save 16% Pay once and save $98 versus $597 on the 3-month plan. $199 /month for 3 months $597 in total, with the same lifetime access. Build Two production agents from scratch: reasoning loops, tools, MCP, multi-agent Format 35 lessons with evals, deployment, and CI/CD baked in Outcome Certification backed by a working agent portfolio 5.0 (26 reviews) Unlock Lifetime Access Start the 3-month plan Try 7 Lessons Free", + "heading": "Course Curriculum", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-0254794793b58e102637f2bd", + "text": "5 Lesson 24: Human-in-the-Loop Through MCP Servers" + }, + { + "span_id": "span-d30fad35a8b42bc39bcdada1", + "text": "6 Lesson 25: Orchestrate and Integrate Our Capstone Agents" + }, + { + "span_id": "span-e90097f11c6da6be272d4a6c", + "text": "7 Quick Quiz 6" + }, + { + "span_id": "span-99279777d81f6c9215393028", + "text": "8 Lesson 26: End-to-End Demo: Generating a Course Lesson" + }, + { + "span_id": "span-ebddd1577a4fad0fe6fe83a7", + "text": "05 Part 3: Evaluation, Observability, Optimizations, and Deployment 10 items + 1 Lesson 27: Agent Observability with Opik" + }, + { + "span_id": "span-b194ee66c24cee0ce12303c3", + "text": "2 Lesson 28: Creating Datasets for AI Evals" + }, + { + "span_id": "span-ccbd0e2cbd20d07bc647750e", + "text": "3 Lesson 29: Defining the Evaluation Processes and Metrics Theory" + }, + { + "span_id": "span-f6de0588f6134dec39bc2eb4", + "text": "4 Lesson 30: Evaluating the Writing Workflow" + }, + { + "span_id": "span-0fd70ca180e7ee2d2114098d", + "text": "5 Quick Quiz 7" + }, + { + "span_id": "span-30d6703f15490c25808471f3", + "text": "6 Lesson 31: Continuous Integration for AI Engineering" + }, + { + "span_id": "span-eb9ceb842c873f495f687d93", + "text": "7 Lesson 32: Preparing For Deployment: Authentication and Docker" + }, + { + "span_id": "span-f68df60646136f9e9937afa9", + "text": "8 Lesson 33: Preparing For Deployment: Database and File Download/Upload" + }, + { + "span_id": "span-28dd9d414fcf5484faadf379", + "text": "9 Lesson 34: Continuous Deployment for AI Engineering" + }, + { + "span_id": "span-8fbace7e66a290dc5d0740bb", + "text": "10 Quick Quiz 8" + }, + { + "span_id": "span-8423481ffcca991da683d6bb", + "text": "06 Part 4: Instructions for Students Capstone Project and Certification 3 items + 1 Part 4: Building Your Own MCP Server (To Receive Certification)" + }, + { + "span_id": "span-c1c58997fd4505047536bea7", + "text": "2 Part 4: Project Submission" + }, + { + "span_id": "span-49afe57e17db4488b49a4191", + "text": "3 Share your success and showcase your new skills!" + }, + { + "span_id": "span-7bb21d31d1aa5fcd7d526d4a", + "text": "07 Extra tools 1 item + 1 AI Tutor - Ask any question!" + }, + { + "span_id": "span-d81f6d5fef1ecf68e4d71e6a", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-f2c5c3a614461d3c910d2888", + "text": "$597 $499 one-time Save 16%" + }, + { + "span_id": "span-e10f69c3dda752408cceb8dd", + "text": "Pay once and save $98 versus $597 on the 3-month plan." + }, + { + "span_id": "span-19f8a85ca1c19a2ee4b3b1b7", + "text": "$199 /month for 3 months" + }, + { + "span_id": "span-a0090734a947e982b8e3d5a2", + "text": "$597 in total, with the same lifetime access." + }, + { + "span_id": "span-b96edccce529423750d58d6f", + "text": "Two production agents from scratch: reasoning loops, tools, MCP, multi-agent" + }, + { + "span_id": "span-795f76d6f3677acd604aeaa2", + "text": "35 lessons with evals, deployment, and CI/CD baked in" + }, + { + "span_id": "span-9af30d2e71288c4b356e2cd8", + "text": "Certification backed by a working agent portfolio" + }, + { + "span_id": "span-88f266cecf77fab1e50b1bed", + "text": "5.0 (26 reviews)" + }, + { + "span_id": "span-bd229d3fa7f914420df1a833", + "text": "Unlock Lifetime Access Start the 3-month plan Try 7 Lessons Free" + } + ] + }, + { + "chunk_id": "tai-d4091d5ce960ba562c7ee566", + "text": "How It Works\nSelf-paced project-based learning designed for working professionals. 01\n\nSelf-Paced + Instructor Support\nLearn on your schedule with lifetime access. Join dedicated Discord channels with instructors and fellow course takers. 02\n\nLearn by Building\nHands-on projects from day one. Build two production agents from scratch, not toy examples or follow-along tutorials. 03\n\nActive Discord Community\nGet unstuck fast. Connect with fellow builders, share progress, and collaborate on projects in an active engineering community. 04\n\nCertificate Upon Completion\nProve you built and deployed production AI. Graduate with certification backed by two real systems and portfolio-ready proof you ship. 05\n\nLifetime Access + Updates\nAll current content plus future updates as AI evolves. Your investment stays valuable. 06\n\nMoney-Back Guarantee\n30-day full refund, no questions asked. Risk-free enrollment.\n\nRisk-Free Learning: 30-Day Money-Back Guarantee\nIf you explore the early part of the course and feel it’s not what you need, you can request a refund within 30 days by emailing us at louis@towardsai.net Start Risk-Free\n\nWho This Course Is For\nFor software engineers, data engineers, data scientists transitioning into AI Engineering 01 Python: Comfortable with functions, classes, and API calls 02 LLMs: Familiar with OpenAI/Claude APIs and basic prompting 03 Docker & Cloud: Understand containerization and deployment basics 04 Skill Level: Intermediate to advanced, expect a challenge 05 Mindset: Learn by building, not watching tutorials\n\nBrought to you by\nTowards AI Towards AI's mission is to make \"building with AI\" more accessible. Since 2019, we have helped teach over 500k about AI with our courses, blogs, tutorials, books, community, and personalized consultancy. Co-instructor", + "heading": "Who This Course Is For", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-47b2a94993f231ba6979b2f3", + "text": "Self-paced project-based learning designed for working professionals." + }, + { + "span_id": "span-f446a8ce701c99570ede9a64", + "text": "Learn on your schedule with lifetime access." + }, + { + "span_id": "span-4fa6c373b79af87745974d2e", + "text": "Join dedicated Discord channels with instructors and fellow course takers." + }, + { + "span_id": "span-3dc767b15104755181689e37", + "text": "Hands-on projects from day one." + }, + { + "span_id": "span-f5982fdc19c5b78e0769a4bb", + "text": "Build two production agents from scratch, not toy examples or follow-along tutorials." + }, + { + "span_id": "span-ff2646ad352f9efd44a3d95b", + "text": "Get unstuck fast." + }, + { + "span_id": "span-1f667c7771283fa12898e0e2", + "text": "Connect with fellow builders, share progress, and collaborate on projects in an active engineering community." + }, + { + "span_id": "span-e7a96d99cde95c60b3ab812b", + "text": "Prove you built and deployed production AI." + }, + { + "span_id": "span-dcccd57facc651f54ad03175", + "text": "Graduate with certification backed by two real systems and portfolio-ready proof you ship." + }, + { + "span_id": "span-ce7899533bec58e8167e7caa", + "text": "All current content plus future updates as AI evolves." + }, + { + "span_id": "span-7cb84ee780ff056c43c2d6ef", + "text": "Your investment stays valuable." + }, + { + "span_id": "span-8417f5af6a545a3b7a089728", + "text": "30-day full refund, no questions asked." + }, + { + "span_id": "span-caf9ec677c454f04692b19bc", + "text": "Risk-free enrollment." + }, + { + "span_id": "span-be386297bec8f01fd1a7371f", + "text": "Risk-Free Learning: 30-Day Money-Back Guarantee" + }, + { + "span_id": "span-551050f0b90217f716310ef5", + "text": "If you explore the early part of the course and feel it’s not what you need, you can request a refund within 30 days by emailing us at louis@towardsai.net" + }, + { + "span_id": "span-923d88cb5ff5ab9375443511", + "text": "Start Risk-Free" + }, + { + "span_id": "span-1c07a31e2f27df69f2996adb", + "text": "For software engineers, data engineers, data scientists transitioning into AI Engineering" + }, + { + "span_id": "span-7cba5e314f78fb042bc26ac5", + "text": "01 Python: Comfortable with functions, classes, and API calls" + }, + { + "span_id": "span-887e79e7ed6fd5b7d399543e", + "text": "02 LLMs: Familiar with OpenAI/Claude APIs and basic prompting" + }, + { + "span_id": "span-7ab64bcb4fa4d9e46b995886", + "text": "03 Docker & Cloud: Understand containerization and deployment basics" + }, + { + "span_id": "span-26a7f60ea63887cca0bccbe3", + "text": "04 Skill Level: Intermediate to advanced, expect a challenge" + }, + { + "span_id": "span-43f1ffb390b05776b026fa8f", + "text": "05 Mindset: Learn by building, not watching tutorials" + }, + { + "span_id": "span-239e50f113785d979a4ebac0", + "text": "Towards AI Towards AI's mission is to make \"building with AI\" more accessible." + }, + { + "span_id": "span-130798b04e450e30731cf7e5", + "text": "Since 2019, we have helped teach over 500k about AI with our courses, blogs, tutorials, books, community, and personalized consultancy." + }, + { + "span_id": "span-94dbd2c65083e8776d430dce", + "text": "Co-instructor" + } + ] + }, + { + "chunk_id": "tai-faae86b9c47828980ee81060", + "text": "Paul Iusztin\nAfter shipping 21 AI apps, Paul Iusztin has used his best-selling LLM Engineer's Handbook, Decoding AI Magazine, and social media platforms to lead 162,000+ AI engineers out of \"demo purgatory\" and into production-grade engineering.\n\nBuild Your First Agent This Week\nStop calling LLMs. Start engineering agents. Own two agents, fully production-ready Gain portfolio-ready skills with certification Master fundamentals that outlast trends Enroll Now Self-paced learning · Regular updates · Risk-free enrollment", + "heading": "Build Your First Agent This Week", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-a26c32f20b81928d95270ffd", + "text": "After shipping 21 AI apps, Paul Iusztin has used his best-selling LLM Engineer's Handbook, Decoding AI Magazine, and social media platforms to lead 162,000+ AI engineers out of \"demo purgatory\" and into production-grade engineering." + }, + { + "span_id": "span-bfb450e6164e84fa1e9e61fa", + "text": "Stop calling LLMs." + }, + { + "span_id": "span-5d4c9c4b1614f09b5bf4d316", + "text": "Start engineering agents." + }, + { + "span_id": "span-a4e363cb699bb9d6923d81d6", + "text": "Own two agents, fully production-ready" + }, + { + "span_id": "span-8fc1a882203bb74cf0eb7245", + "text": "Gain portfolio-ready skills with certification" + }, + { + "span_id": "span-088d0b4100baf76931a434e0", + "text": "Master fundamentals that outlast trends" + }, + { + "span_id": "span-3afd3cdd033ed5fe1627b023", + "text": "Enroll Now" + }, + { + "span_id": "span-42d7e3a04746b5208acbde44", + "text": "Self-paced learning · Regular updates · Risk-free enrollment" + } + ] + }, + { + "chunk_id": "tai-db7c79a374574a9416d10c3a", + "text": "FAQ\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. How does it compare to your \"Full Stack AI Engineering\" course? + The Full Stack AI Engineering course teaches the complete LLM product stack including RAG, fine-tuning, deployment, and agents through one central AI product. The Agents course goes much deeper into designing, orchestrating, evaluating, and deploying production-grade autonomous systems. If Full Stack teaches you how to build an AI product, Agents teaches you how to engineer the autonomy inside it. Compare the Full Stack course . What projects will I work on? + You'll build two interconnected production agents from the ground up: a Research Agent and a Writing Workflow, with human-in-the-loop capabilities, while mastering the foundations of agentic architecture. How is this course different from other AI courses? + Unlike many theoretical courses, this one is fully hands-on, project-based, and focused on the skills top companies are hiring for right now. We guide you through every step, ensuring you not only understand the concepts but can also apply them immediately. Do you have a referral or affiliate program? + For every referral you provide, you’ll gain access to our exclusive eBook. If you refer three or more people, we’ll send you a physical copy of our book as a thank you! Additionally, we have an affiliate program for individuals with an audience. By joining, you can earn commissions for every successful referral made through your unique affiliate link. Please email Louis-François at louis@towardsai.net with proof of referral. Can I get a discount? + Yes, there are three ways to receive a discount on our courses. If you're a student, we offer a 50% discount on all courses. Additionally,", + "heading": "FAQ", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-8141c21df12387b1ae8098bf", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-dd8cc419912db50c1cb22a68", + "text": "You'll build two interconnected production agents from the ground up: a Research Agent and a Writing Workflow, with human-in-the-loop capabilities, while mastering the foundations of agentic architecture." + }, + { + "span_id": "span-acba9bc4635613b031ac3e5a", + "text": "Unlike many theoretical courses, this one is fully hands-on, project-based, and focused on the skills top companies are hiring for right now." + }, + { + "span_id": "span-55bb6886847edde29bc3f207", + "text": "We guide you through every step, ensuring you not only understand the concepts but can also apply them immediately." + } + ] + }, + { + "chunk_id": "tai-1508a8684b292045012ce6ab", + "text": "FAQ\nif you've previously purchased our book or another course, you're eligible for a discount. We also provide a bundle discount for groups of two or more people. For further details or to claim your discount, please email Louis-François at louis@towardsai.net for a follow-up. Can I take this course within my company? + Yes! We offer both course bundles and custom training solutions tailored specifically for companies. For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net . Our refund policy + If you feel the course isn’t right for you, we’ll provide a 100% refund within 30 days—no hassle, no worries.", + "heading": "FAQ", + "index": 10, + "evidence_spans": [ + { + "span_id": "span-a2249eb144d97ad4738945cf", + "text": "We offer both course bundles and custom training solutions tailored specifically for companies." + }, + { + "span_id": "span-4bf1482d730513a9ae91b6d6", + "text": "For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net ." + }, + { + "span_id": "span-75c218be0776aaa47004eb23", + "text": "Our refund policy + If you feel the course isn’t right for you, we’ll provide a 100% refund within 30 days—no hassle, no worries." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "8e3f805a4bc7c547582d9f00673eed0c412db4047b49854cd13306450d469367", + "content_hash": "8e3f805a4bc7c547582d9f00673eed0c412db4047b49854cd13306450d469367", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/agent-engineering/" + ], + "evidence_hash": "8b4b625d29bb04ecfb32904c85bb66e646af449102e5341c065637aa2f842eee" + }, + { + "discovered_url": "https://towardsai.com/academy/llm-primer/", + "url": "https://towardsai.com/academy/llm-primer/", + "canonical_url": "https://towardsai.com/academy/llm-primer/", + "host": "towardsai.com", + "path": "/academy/llm-primer", + "kind": "course", + "offer_id": "llm-primer", + "entity_id": "offer:llm-primer", + "title": "10-Hour LLM Fundamentals: Video Crash Course | Towards AI Academy", + "review_title": "10-Hour LLM Fundamentals", + "reviewed_summary": "Focused LLM fundamentals video course.", + "meta_description": "A 10-hour video crash course on LLM fundamentals: prompting, RAG, evaluation, agents, and optimization. Know what to use, when, and why. Lifetime access for $199.", + "headings": [ + "10 hours to LLM fundamentals", + "When to prompt, RAG, fine-tune, or deploy agents", + "Know what to use, when, and why", + "Customisation, retrieval, evaluation", + "Deploy with confidence", + "What our students say", + "Five sessions, prompt to production", + "Who is this course for?", + "Launch AI features that actually ship", + "Lead with confidence", + "Future-proof your career", + "Frequently asked questions", + "Your AI skillset, 10x'd in 10 hours" + ], + "text": "Video crash course · Self-paced\n\n10 hours to LLM fundamentals\nUnderstand, build, evaluate, automate, and maintain robust LLM solutions. Five focused video sessions take you from a basic prompt to a full production rollout. See the sessions Taught by Louis-François Bouchard Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught 5 in-depth 2-hour video sessions $199 one-time, down from $299 100% video, built for busy professionals Pay once 3 monthly payments $299 $199 one-time Save 17% Pay once and save $41 versus $240 on the 3-month plan. $80 /month for 3 months $240 in total, with the same lifetime access. Learn When to prompt, RAG, fine-tune, or deploy agents, and why Format Five in-depth 2-hour video sessions, built for busy professionals Outcome Certification and lifetime access, with all future updates Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 The decision skill\n\nWhen to prompt, RAG, fine-tune, or deploy agents\nThe real challenge isn't coding; it's knowing where to begin. This course gives you a mental map of the LLM landscape so you know what to try first. Use LLMs for your domain\n\nKnow what to use, when, and why\nUnderstand how these models work, how to compare them, and when to use them. Practical case studies on agents and domain-specific projects help you choose the best solution for a given scenario. The core developer toolkit\n\nCustomisation, retrieval, evaluation\nAdopt databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows: the core skills for customising LLMs to specific industries and improving reliability. Ideas into products\n\nDeploy with confidence\nDevelop intuition for techniques like distillation, quantization, RLHF, and defences against prompt-hacking, so your deployments hold up at scale. 02 Student outcomes\n\nWhat our students say\n03 The live syllabus\n\nFive sessions, prompt to production\nEach 2-hour session moves you one layer deeper, with quizzes to lock in the concepts. 01 Foundational AI knowledge and using LLMs session 1 + 1 Foundational Knowledge and Using LLMs Free preview 2 Quiz 02 Designing end-to-end systems session 2 + 1 Building on Top of LLMs Free preview 2 Quiz 03 Evaluating RAG and LLM pipelines session 3 + 1 Evaluations 2 Quiz 04 AI workflows and agents, with case studies session 4 + 1 Orchestrating Agent-Based Solutions 2 Quiz 05 Optimization, guardrails, and advanced techniques session 5 + 1 Guardrails, Advanced Techniques, Optimizations and Monitoring 2 Quiz 06 Re-training models bonus + 1 Re-training Models 2 Share your success and showcase your new skills! 3 Quiz Updated regularly. Certification, community support, and lifetime access included. Pay once 3 monthly payments $299 $199 one-time Save 17% Pay once and save $41 versus $240 on the 3-month plan. $80 /month for 3 months $240 in total, with the same lifetime access. Learn When to prompt, RAG, fine-tune, or deploy agents, and why Format Five in-depth 2-hour video sessions, built for busy professionals Outcome Certification and lifetime access, with all future updates Community support and AI tutor Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 04 Made for busy professionals\n\nWho is this course for?\nDesigned as a one-day bootcamp for software professionals, language agnostic, and relevant well beyond developers. No prior AI knowledge required; a Python foundation helps but isn't essential. Ship features that hold up\n\nLaunch AI features that actually ship\nEvaluate quality, integrate domain data, and monitor for safety, so your features hold up in production rather than only in a demo. Make the big calls\n\nLead with confidence\nWhether you're guiding a team, building a startup, or upskilling, develop the mental models to decide when to fine-tune, how to scale, and where to push back. A skillset in demand\n\nFuture-proof your career\nIn five focused sessions, gain a skillset already in demand across product, engineering, and research teams, from startups to enterprise. + Ready to go deeper? Continue with Full Stack AI Engineering , our flagship code-heavy program, and ask us about bundle pricing. + Not a coder? Master AI for Work covers the no-code path for professionals. 05 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. How long does this course take to complete? + It is self-paced; the average completion time is 10 hours across five 2-hour sessions. Do I need a technical background? + You don't need a specific programming language. The course is designed for software professionals who want the most out of LLM development techniques, and prepares you to work in the language of your choice. Python has the broadest compatibility in the LLM ecosystem. Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. Can I take this course within my company? + Yes, we partner with companies to offer this course to teams and can customise the content. See our enterprise programme or get in touch . Lifetime access · $199\n\nYour AI skillset, 10x'd in 10 hours\nWhether you're solo or leading a team, learn to build safer, faster, smarter LLM systems, on time and on budget. Level up in 5 sessions Compare the full program", + "links": [ + { + "text": "Unlock lifetime access", + "url": "https://academy.towardsai.net/cart/add_product/3399532?price_id=4310653" + }, + { + "text": "Start the 3-month plan", + "url": "https://academy.towardsai.net/enroll/3399532?price_id=4813877" + }, + { + "text": "Foundational Knowledge and Using LLMs Free preview", + "url": "https://academy.towardsai.net/enroll/3399532?et=free_trial" + }, + { + "text": "Building on Top of LLMs Free preview", + "url": "https://academy.towardsai.net/enroll/3399532?et=free_trial" + }, + { + "text": "Full Stack AI Engineering", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Master AI for Work", + "url": "https://towardsai.com/academy/ai-for-work/" + }, + { + "text": "Contact us", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "enterprise programme", + "url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/" + }, + { + "text": "get in touch", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "Level up in 5 sessions", + "url": "https://academy.towardsai.net/cart/add_product/3399532?price_id=4310653" + }, + { + "text": "Compare the full program", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + } + ], + "chunks": [ + { + "chunk_id": "tai-013653705151f90a7e88987e", + "text": "Video crash course · Self-paced\n\n10 hours to LLM fundamentals\nUnderstand, build, evaluate, automate, and maintain robust LLM solutions. Five focused video sessions take you from a basic prompt to a full production rollout. See the sessions Taught by Louis-François Bouchard Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught 5 in-depth 2-hour video sessions $199 one-time, down from $299 100% video, built for busy professionals Pay once 3 monthly payments $299 $199 one-time Save 17% Pay once and save $41 versus $240 on the 3-month plan. $80 /month for 3 months $240 in total, with the same lifetime access. Learn When to prompt, RAG, fine-tune, or deploy agents, and why Format Five in-depth 2-hour video sessions, built for busy professionals Outcome Certification and lifetime access, with all future updates Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 The decision skill\n\nWhen to prompt, RAG, fine-tune, or deploy agents\nThe real challenge isn't coding; it's knowing where to begin. This course gives you a mental map of the LLM landscape so you know what to try first. Use LLMs for your domain\n\nKnow what to use, when, and why\nUnderstand how these models work, how to compare them, and when to use them. Practical case studies on agents and domain-specific projects help you choose the best solution for a given scenario. The core developer toolkit\n\nCustomisation, retrieval, evaluation\nAdopt databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows: the core skills for customising LLMs to specific industries and improving reliability. Ideas into products", + "heading": "10 hours to LLM fundamentals", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-7d1687cac72c103110c0d139", + "text": "Video crash course · Self-paced" + }, + { + "span_id": "span-dc6925c5ec63b5a99e0235b9", + "text": "10 hours to LLM fundamentals" + }, + { + "span_id": "span-65df0b28735d89ab8ad6501c", + "text": "Understand, build, evaluate, automate, and maintain robust LLM solutions." + }, + { + "span_id": "span-326f8c0857415a78b90a221b", + "text": "Five focused video sessions take you from a basic prompt to a full production rollout." + }, + { + "span_id": "span-7cf367b1e9dc975921bb9c28", + "text": "See the sessions" + }, + { + "span_id": "span-96e41c21b21da0d50ba35208", + "text": "Taught by Louis-François Bouchard Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught" + }, + { + "span_id": "span-0f9dcc4b2f1602bf414df8a6", + "text": "5 in-depth 2-hour video sessions" + }, + { + "span_id": "span-e685e738c5ddbd67032f77c8", + "text": "$199 one-time, down from $299" + }, + { + "span_id": "span-c2ec338ceb7d35a43b37105f", + "text": "100% video, built for busy professionals" + }, + { + "span_id": "span-ed9333aee8d10bb6e0b913a3", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-74df0c1f69957fefb44d5a15", + "text": "$299 $199 one-time Save 17%" + }, + { + "span_id": "span-078aa80484d4f08326b3cbc6", + "text": "Pay once and save $41 versus $240 on the 3-month plan." + }, + { + "span_id": "span-a2f2d1c270f0bdfc797c7905", + "text": "$80 /month for 3 months" + }, + { + "span_id": "span-afd68c2dd61af908713e9634", + "text": "$240 in total, with the same lifetime access." + }, + { + "span_id": "span-609200ca1e3b4bce18866ee2", + "text": "When to prompt, RAG, fine-tune, or deploy agents, and why" + }, + { + "span_id": "span-f8c2ca25798e515bdf8290b0", + "text": "Five in-depth 2-hour video sessions, built for busy professionals" + }, + { + "span_id": "span-dfe37c60fa3c0aa3c1abcb36", + "text": "Certification and lifetime access, with all future updates" + }, + { + "span_id": "span-0ac71735710de186628bf856", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-63962c200f30a6994385a6f9", + "text": "01 The decision skill" + }, + { + "span_id": "span-5c8889e19f7b1917e9d1e823", + "text": "The real challenge isn't coding; it's knowing where to begin." + }, + { + "span_id": "span-a13a576e0a926e8dc82f99b0", + "text": "This course gives you a mental map of the LLM landscape so you know what to try first." + }, + { + "span_id": "span-c4c9bc7bfae4223d5cab6170", + "text": "Use LLMs for your domain" + }, + { + "span_id": "span-942bcdcedd71d213ec520575", + "text": "Understand how these models work, how to compare them, and when to use them." + }, + { + "span_id": "span-ba94979ed3968eb8c4c84c86", + "text": "Practical case studies on agents and domain-specific projects help you choose the best solution for a given scenario." + }, + { + "span_id": "span-174c677975abec1c147467c5", + "text": "The core developer toolkit" + }, + { + "span_id": "span-6b718a940a7e7e144789489a", + "text": "Adopt databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows: the core skills for customising LLMs to specific industries and improving reliability." + }, + { + "span_id": "span-9e6e6becc79dceb9db16d762", + "text": "Ideas into products" + } + ] + }, + { + "chunk_id": "tai-7349eef88d25a19b2cea6847", + "text": "Deploy with confidence\nDevelop intuition for techniques like distillation, quantization, RLHF, and defences against prompt-hacking, so your deployments hold up at scale. 02 Student outcomes\n\nWhat our students say\n03 The live syllabus\n\nFive sessions, prompt to production\nEach 2-hour session moves you one layer deeper, with quizzes to lock in the concepts. 01 Foundational AI knowledge and using LLMs session 1 + 1 Foundational Knowledge and Using LLMs Free preview 2 Quiz 02 Designing end-to-end systems session 2 + 1 Building on Top of LLMs Free preview 2 Quiz 03 Evaluating RAG and LLM pipelines session 3 + 1 Evaluations 2 Quiz 04 AI workflows and agents, with case studies session 4 + 1 Orchestrating Agent-Based Solutions 2 Quiz 05 Optimization, guardrails, and advanced techniques session 5 + 1 Guardrails, Advanced Techniques, Optimizations and Monitoring 2 Quiz 06 Re-training models bonus + 1 Re-training Models 2 Share your success and showcase your new skills! 3 Quiz Updated regularly. Certification, community support, and lifetime access included. Pay once 3 monthly payments $299 $199 one-time Save 17% Pay once and save $41 versus $240 on the 3-month plan. $80 /month for 3 months $240 in total, with the same lifetime access. Learn When to prompt, RAG, fine-tune, or deploy agents, and why Format Five in-depth 2-hour video sessions, built for busy professionals Outcome Certification and lifetime access, with all future updates Community support and AI tutor Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 04 Made for busy professionals", + "heading": "Five sessions, prompt to production", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-910f094d305bf7fd1ca64d5f", + "text": "Develop intuition for techniques like distillation, quantization, RLHF, and defences against prompt-hacking, so your deployments hold up at scale." + }, + { + "span_id": "span-3bc700250f3d07fe286acb58", + "text": "02 Student outcomes" + }, + { + "span_id": "span-a9b71cc0aec7b3c7f085a666", + "text": "03 The live syllabus" + }, + { + "span_id": "span-846ac04750e2b718376a5f85", + "text": "Each 2-hour session moves you one layer deeper, with quizzes to lock in the concepts." + }, + { + "span_id": "span-9d1822e71df1acc80cd4e857", + "text": "01 Foundational AI knowledge and using LLMs session 1 + 1 Foundational Knowledge and Using LLMs Free preview" + }, + { + "span_id": "span-fcf55ad4531c7d0e153eb9da", + "text": "2 Quiz" + }, + { + "span_id": "span-1fc82841aaa1bffb15e509b6", + "text": "02 Designing end-to-end systems session 2 + 1 Building on Top of LLMs Free preview" + }, + { + "span_id": "span-c994a50379db86064be888e9", + "text": "03 Evaluating RAG and LLM pipelines session 3 + 1 Evaluations" + }, + { + "span_id": "span-da6a7cf862877e39dcb68751", + "text": "04 AI workflows and agents, with case studies session 4 + 1 Orchestrating Agent-Based Solutions" + }, + { + "span_id": "span-c3942ba9b9fd2f12ced377f2", + "text": "05 Optimization, guardrails, and advanced techniques session 5 + 1 Guardrails, Advanced Techniques, Optimizations and Monitoring" + }, + { + "span_id": "span-814bebe5bbb3c76a6c764262", + "text": "06 Re-training models bonus + 1 Re-training Models" + }, + { + "span_id": "span-5e14a74128988d936003399e", + "text": "2 Share your success and showcase your new skills!" + }, + { + "span_id": "span-5d8651e1e756cdd1a0c0bc83", + "text": "3 Quiz" + }, + { + "span_id": "span-9d839c1211d498a2b1c56727", + "text": "Updated regularly." + }, + { + "span_id": "span-afb35dec53cc247d6813b036", + "text": "Certification, community support, and lifetime access included." + }, + { + "span_id": "span-7a1872c5aa436dc061c76356", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-c5e4bd5d39552690265bbd63", + "text": "$299 $199 one-time Save 17%" + }, + { + "span_id": "span-21727e6c4baefd4bc244434c", + "text": "Pay once and save $41 versus $240 on the 3-month plan." + }, + { + "span_id": "span-e3ae229f91adb84b4e281ac9", + "text": "$80 /month for 3 months" + }, + { + "span_id": "span-b2fcfdb8b01c25b75b2803df", + "text": "$240 in total, with the same lifetime access." + }, + { + "span_id": "span-b242b977bfc6bdf1d9a45b0e", + "text": "When to prompt, RAG, fine-tune, or deploy agents, and why" + }, + { + "span_id": "span-1a4acbfe3174b10e110d5b55", + "text": "Five in-depth 2-hour video sessions, built for busy professionals" + }, + { + "span_id": "span-30ca02dd9b0d521e00a4e824", + "text": "Certification and lifetime access, with all future updates" + }, + { + "span_id": "span-0d9b9876d0a923937cee3af0", + "text": "Community support and AI tutor" + }, + { + "span_id": "span-581e4b44901f8df467da2de6", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-d87f3183eff47e26203ebebc", + "text": "04 Made for busy professionals" + } + ] + }, + { + "chunk_id": "tai-f9c2508eac63d6bf53d9c49c", + "text": "Who is this course for?\nDesigned as a one-day bootcamp for software professionals, language agnostic, and relevant well beyond developers. No prior AI knowledge required; a Python foundation helps but isn't essential. Ship features that hold up\n\nLaunch AI features that actually ship\nEvaluate quality, integrate domain data, and monitor for safety, so your features hold up in production rather than only in a demo. Make the big calls\n\nLead with confidence\nWhether you're guiding a team, building a startup, or upskilling, develop the mental models to decide when to fine-tune, how to scale, and where to push back. A skillset in demand\n\nFuture-proof your career\nIn five focused sessions, gain a skillset already in demand across product, engineering, and research teams, from startups to enterprise. + Ready to go deeper? Continue with Full Stack AI Engineering , our flagship code-heavy program, and ask us about bundle pricing. + Not a coder? Master AI for Work covers the no-code path for professionals. 05 Questions", + "heading": "Future-proof your career", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-968b5403510ce45472bb1ebd", + "text": "Designed as a one-day bootcamp for software professionals, language agnostic, and relevant well beyond developers." + }, + { + "span_id": "span-a3bd6d990c4b2a400e2f00d0", + "text": "No prior AI knowledge required; a Python foundation helps but isn't essential." + }, + { + "span_id": "span-1a637c42671fbaa2ed0677c7", + "text": "Ship features that hold up" + }, + { + "span_id": "span-f50287d2397bae7d0a69c840", + "text": "Evaluate quality, integrate domain data, and monitor for safety, so your features hold up in production rather than only in a demo." + }, + { + "span_id": "span-35e88a2d30fd8b8a3aa86f48", + "text": "Make the big calls" + }, + { + "span_id": "span-bd11d03ad644a71303420945", + "text": "Whether you're guiding a team, building a startup, or upskilling, develop the mental models to decide when to fine-tune, how to scale, and where to push back." + }, + { + "span_id": "span-ea66db0f42a108bcdf7c4cca", + "text": "A skillset in demand" + }, + { + "span_id": "span-0cc6af3cfa7d7d17fb6398c3", + "text": "In five focused sessions, gain a skillset already in demand across product, engineering, and research teams, from startups to enterprise." + }, + { + "span_id": "span-bb34de6c8bcb281fea51fa49", + "text": "+ Ready to go deeper?" + }, + { + "span_id": "span-594fadb63999016053ed8c9d", + "text": "Continue with Full Stack AI Engineering , our flagship code-heavy program, and ask us about bundle pricing." + }, + { + "span_id": "span-faa0123e3211e592f3435973", + "text": "+ Not a coder?" + }, + { + "span_id": "span-dc1b0eee3bd8a48404cc5826", + "text": "Master AI for Work covers the no-code path for professionals." + }, + { + "span_id": "span-7549cc1536ac7224462e59ad", + "text": "05 Questions" + } + ] + }, + { + "chunk_id": "tai-d8b1b509c50970a14ab67bff", + "text": "Frequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. How long does this course take to complete? + It is self-paced; the average completion time is 10 hours across five 2-hour sessions. Do I need a technical background? + You don't need a specific programming language. The course is designed for software professionals who want the most out of LLM development techniques, and prepares you to work in the language of your choice. Python has the broadest compatibility in the LLM ecosystem. Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. Can I take this course within my company? + Yes, we partner with companies to offer this course to teams and can customise the content. See our enterprise programme or get in touch . Lifetime access · $199\n\nYour AI skillset, 10x'd in 10 hours\nWhether you're solo or leading a team, learn to build safer, faster, smarter LLM systems, on time and on budget. Level up in 5 sessions Compare the full program", + "heading": "Frequently asked questions", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-bf4a981de7e731d89e21c445", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-ec76f972be9654cd233cb94e", + "text": "It is self-paced; the average completion time is 10 hours across five 2-hour sessions." + }, + { + "span_id": "span-74f4cea04655401e6e5e729a", + "text": "You don't need a specific programming language." + }, + { + "span_id": "span-edc18743ee143ee5de2df77b", + "text": "The course is designed for software professionals who want the most out of LLM development techniques, and prepares you to work in the language of your choice." + }, + { + "span_id": "span-ad7014b52a30082c07470fc9", + "text": "Python has the broadest compatibility in the LLM ecosystem." + }, + { + "span_id": "span-58c00bb2d452020400a2dff6", + "text": "The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves." + }, + { + "span_id": "span-6c32763665172ab74ebb8303", + "text": "Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing." + }, + { + "span_id": "span-4f34307d27b5a3d56fe5ea82", + "text": "Contact us to claim." + }, + { + "span_id": "span-9f646bbc7fdb511313cbaaa7", + "text": "Yes, we partner with companies to offer this course to teams and can customise the content." + }, + { + "span_id": "span-0e703737fc47a12302bfc58b", + "text": "See our enterprise programme or get in touch ." + }, + { + "span_id": "span-9c4f0d2e2f35587fb2e2209f", + "text": "Lifetime access · $199" + }, + { + "span_id": "span-0094648c74f29f94b1f03bf7", + "text": "Your AI skillset, 10x'd in 10 hours" + }, + { + "span_id": "span-3e4654583ad0846a752630bc", + "text": "Whether you're solo or leading a team, learn to build safer, faster, smarter LLM systems, on time and on budget." + }, + { + "span_id": "span-ff0dd73e67d10bb457881e3e", + "text": "Level up in 5 sessions Compare the full program" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "a083e8e136346b63103a30dff06126dfa5cfe07e4243488b14a18383f8177a93", + "content_hash": "a083e8e136346b63103a30dff06126dfa5cfe07e4243488b14a18383f8177a93", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/llm-primer/" + ], + "evidence_hash": "99a645c0b976eec560f72fc2de22f7ccb9578cc4170f516190104041780ed118" + }, + { + "discovered_url": "https://towardsai.com/academy/python-for-ai-engineering/", + "url": "https://towardsai.com/academy/python-for-ai-engineering/", + "canonical_url": "https://towardsai.com/academy/python-for-ai-engineering/", + "host": "towardsai.com", + "path": "/academy/python-for-ai-engineering", + "kind": "course", + "offer_id": "python-for-ai-engineering", + "entity_id": "offer:python-for-ai-engineering", + "title": "Beginner Python for AI Engineering: The LLM-Native Python Course | Towards AI Academy", + "review_title": "Python for AI Engineering", + "reviewed_summary": "Beginner Python foundations for AI engineering.", + "meta_description": "The first LLM-native Python course for complete beginners. Learn Python by building AI projects with coding assistants, from first script to a RAG capstone. Lifetime access for $149.", + "headings": [ + "The first LLM-native Python course", + "Don't just interact with LLMs. Build with them.", + "Python tailored for LLM work", + "Hands-on with LLM tools and APIs", + "Learn with AI assistants, deliberately", + "What's stopping you from learning Python?", + "What our students say", + "Course curriculum", + "Who is this course for?", + "Go beyond prompting", + "Open the door to AI roles", + "Tools that work for you", + "Frequently asked questions", + "Learn Python the LLM-native way" + ], + "text": "Foundations · Complete beginners welcome\n\nThe first LLM-native Python course\nLearn Python the modern way: iteratively, with AI assistants, by building real projects from a budget tracker to a RAG application. No prior coding experience required. Unlock Lifetime Access for $149 with all future updates See the curriculum 30-day money-back guarantee · lifetime access · updated weekly 38 lessons, from first script to capstone 12+ hands-on projects you build yourself 0 prior coding experience required 01 Beyond prompting\n\nDon't just interact with LLMs. Build with them.\nFoundations for AI development\n\nPython tailored for LLM work\nA clear grounding in Python concepts, tools, and best practices aimed specifically at building with and fine-tuning LLMs, so you finish able to build AI applications that matter to your daily tasks and long-term goals. Real tools from day one\n\nHands-on with LLM tools and APIs\nGo beyond syntax: work with libraries like Hugging Face and integrate open-source models into your projects while your Python skills build up alongside. Code the modern way\n\nLearn with AI assistants, deliberately\nUse GitHub Copilot and ChatGPT to write better Python faster. You learn to direct the assistant and check its work rather than copy blindly, which is exactly how professionals code now. 02 Honest answers\n\nWhat's stopping you from learning Python?\n01 \"I don't have months for this.\" Learning with AI assistance compresses the curve: weeks, not months. 02 \"I'm not a math genius.\" Python reads closer to English than to equations. We teach the logic; you don't need the math. 03 \"LLMs can code for me.\" They assist. You still need Python to build products that scale and to catch what the model gets wrong. 04 \"AI keeps changing.\" The course updates weekly, and we teach you how to keep teaching yourself as it evolves. 05 \"What if I get stuck?\" Course tutors on Discord, plus techniques for using AI itself to get unstuck. 06 \"Am I too late?\" The field is still young. Python remains the skill that compounds fastest with it. 03 Student outcomes\n\nWhat our students say\n04 The live syllabus\n\nCourse curriculum\nSix parts, 38 lessons, a dozen projects, and a portfolio-ready capstone. The first lessons of every part are free to preview. 01 Introduction 4 lessons + 1 Course Introduction and Rationale Free preview 2 Course Structure Free preview 3 How LLMs can be used as Coding Assistants Free preview 4 Computer Science Fundamentals Free preview 02 Using LLMs 4 lessons + 1 Part 1 Overview: Using LLMs Free preview 2 Introduction to ChatGPT, Claude, and Gemini 3 Prompting 101 4 Quiz 03 Beginner code projects 8 lessons + 1 Part 2 Overview: Beginner Code Projects Free preview 2 Personal Budget Tracker Free preview 3 Healthy Habits Tracker 4 Quiz 5 Understanding Python Environments and Tools Free preview 6 Analyzing Coffee Shop Sales with Notebooks 7 Text Adventure Game 8 Quiz 04 Intermediate code projects 8 lessons + 1 Part 3 Overview: Intermediate Code Projects Free preview 2 GitHub and Code Versioning 3 Random Password Generator 4 Web Scraping 5 Quiz 6 Interactive Quiz App with Gradio 7 Personal Finance Dashboard with Charts 8 Quiz 05 Code projects using LLMs 8 lessons + 1 Part 4 Overview: Code Projects Using LLMs Free preview 2 How LLMs work 3 News Analyzer: Summarize, Sentiment, and Tags 4 Chat with Your Documents: A RAG Application 5 Create an API for Drafting Social Posts 6 Quiz 7 Future of AI-Driven Programming 8 The Future in Practice: Your Guide to AI Coding Tools 06 Next steps and capstone project 6 lessons + 1 Part 5 Overview: Next Steps and Capstone Project Free preview 2 Where to Go From Here: Paths to Explore 3 Capstone Project Description 4 Capstone Project DropBox 5 Share your success and showcase your new skills! 6 AI Tutor: Ask any question! Updated weekly with new lessons as the field evolves. $149 one-time · lifetime access 38 lessons, 12+ hands-on projects Portfolio-ready capstone project Lifetime access, updated weekly Certification on completion Discord tutors and AI tutor support Unlock lifetime access 30-day money-back guarantee, no questions asked 05 Start from zero\n\nWho is this course for?\nComplete beginners, aspiring AI developers, and professionals who want to lead AI projects. No prior coding, no software background: just curiosity, a computer, and internet access. From black box to toolbox\n\nGo beyond prompting\nMost people just use AI. With Python you start customising, fine-tuning, and building AI solutions that actually fit your needs. The hiring signal\n\nOpen the door to AI roles\nPython is the language of AI. Whether you want to automate tasks, analyse data, or fine-tune models, it's the skill companies look for first. Build, don't wait\n\nTools that work for you\nA chatbot, a research assistant, an automation: with Python you build what you need instead of relying on whatever tool happens to exist. + Next step after this course: Full Stack AI Engineering , where these Python skills become production LLM systems. 06 Questions\n\nFrequently asked questions\nDo I need prior experience with coding or AI? + No. The course teaches Python from scratch and assumes no software background. Will I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What projects will I work on? + A dozen builds from a personal budget tracker to web scraping, a Gradio quiz app, a news analyzer, and a chat-with-your-documents RAG application, ending in a capstone project of your own. Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. What is your refund policy? + If, after completing the first section, you feel the course isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries. Lifetime access · $149\n\nLearn Python the LLM-native way\nCreate with LLMs and transform your work with code, starting from zero. Start learning smarter See the full pathway", + "links": [ + { + "text": "Unlock Lifetime Access for $149 with all future updates", + "url": "https://academy.towardsai.net/cart/add_product/3147320?price_id=4025386" + }, + { + "text": "Course Introduction and Rationale Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Course Structure Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "How LLMs can be used as Coding Assistants Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Computer Science Fundamentals Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Part 1 Overview: Using LLMs Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Part 2 Overview: Beginner Code Projects Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Personal Budget Tracker Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Understanding Python Environments and Tools Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Part 3 Overview: Intermediate Code Projects Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Part 4 Overview: Code Projects Using LLMs Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Part 5 Overview: Next Steps and Capstone Project Free preview", + "url": "https://academy.towardsai.net/enroll/3147320?et=free_trial" + }, + { + "text": "Unlock lifetime access", + "url": "https://academy.towardsai.net/cart/add_product/3147320?price_id=4025386" + }, + { + "text": "Full Stack AI Engineering", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Contact us", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "Start learning smarter", + "url": "https://academy.towardsai.net/cart/add_product/3147320?price_id=4025386" + }, + { + "text": "See the full pathway", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-991560dbd35de5c14b912066", + "text": "Foundations · Complete beginners welcome\n\nThe first LLM-native Python course\nLearn Python the modern way: iteratively, with AI assistants, by building real projects from a budget tracker to a RAG application. No prior coding experience required. Unlock Lifetime Access for $149 with all future updates See the curriculum 30-day money-back guarantee · lifetime access · updated weekly 38 lessons, from first script to capstone 12+ hands-on projects you build yourself 0 prior coding experience required 01 Beyond prompting\n\nDon't just interact with LLMs. Build with them.\nFoundations for AI development\n\nPython tailored for LLM work\nA clear grounding in Python concepts, tools, and best practices aimed specifically at building with and fine-tuning LLMs, so you finish able to build AI applications that matter to your daily tasks and long-term goals. Real tools from day one\n\nHands-on with LLM tools and APIs\nGo beyond syntax: work with libraries like Hugging Face and integrate open-source models into your projects while your Python skills build up alongside. Code the modern way\n\nLearn with AI assistants, deliberately\nUse GitHub Copilot and ChatGPT to write better Python faster. You learn to direct the assistant and check its work rather than copy blindly, which is exactly how professionals code now. 02 Honest answers", + "heading": "The first LLM-native Python course", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-ade3915d33d25d435678976f", + "text": "Foundations · Complete beginners welcome" + }, + { + "span_id": "span-2221d3ebf08ec42f48fb6be1", + "text": "Learn Python the modern way: iteratively, with AI assistants, by building real projects from a budget tracker to a RAG application." + }, + { + "span_id": "span-9650c64548807dc5b9af3d92", + "text": "No prior coding experience required." + }, + { + "span_id": "span-34445bb96f3e64d39b6c9d8e", + "text": "Unlock Lifetime Access for $149 with all future updates See the curriculum" + }, + { + "span_id": "span-d5daa3553d1c27b29277af0a", + "text": "30-day money-back guarantee · lifetime access · updated weekly 38 lessons, from first script to capstone" + }, + { + "span_id": "span-b3e456111d3696c97a7df5c5", + "text": "12+ hands-on projects you build yourself" + }, + { + "span_id": "span-9b96198fe18dfe1cb173897a", + "text": "0 prior coding experience required" + }, + { + "span_id": "span-48b44304164ae2e32d88c526", + "text": "01 Beyond prompting" + }, + { + "span_id": "span-ae9746f702a71c907a7c92cb", + "text": "Foundations for AI development" + }, + { + "span_id": "span-e8ec537be403c1cc5abdf811", + "text": "A clear grounding in Python concepts, tools, and best practices aimed specifically at building with and fine-tuning LLMs, so you finish able to build AI applications that matter to your daily tasks and long-term goals." + }, + { + "span_id": "span-05d4864723f004a477ee5395", + "text": "Real tools from day one" + }, + { + "span_id": "span-e2704481f59b85ba4803c293", + "text": "Go beyond syntax: work with libraries like Hugging Face and integrate open-source models into your projects while your Python skills build up alongside." + }, + { + "span_id": "span-c758841a83d3f87ccab1376e", + "text": "Code the modern way" + }, + { + "span_id": "span-e7be5c7f0b71c922b4704362", + "text": "Use GitHub Copilot and ChatGPT to write better Python faster." + }, + { + "span_id": "span-243824bcc5a3b81fb67a0ee8", + "text": "You learn to direct the assistant and check its work rather than copy blindly, which is exactly how professionals code now." + }, + { + "span_id": "span-31f59d6d4e18ce63944321ee", + "text": "02 Honest answers" + } + ] + }, + { + "chunk_id": "tai-b85ccc03780bb7546d51be23", + "text": "What's stopping you from learning Python?\n01 \"I don't have months for this.\" Learning with AI assistance compresses the curve: weeks, not months. 02 \"I'm not a math genius.\" Python reads closer to English than to equations. We teach the logic; you don't need the math. 03 \"LLMs can code for me.\" They assist. You still need Python to build products that scale and to catch what the model gets wrong. 04 \"AI keeps changing.\" The course updates weekly, and we teach you how to keep teaching yourself as it evolves. 05 \"What if I get stuck?\" Course tutors on Discord, plus techniques for using AI itself to get unstuck. 06 \"Am I too late?\" The field is still young. Python remains the skill that compounds fastest with it. 03 Student outcomes\n\nWhat our students say\n04 The live syllabus", + "heading": "What's stopping you from learning Python?", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-b53f5b0c1bcfdacea08d0448", + "text": "01 \"I don't have months for this.\" Learning with AI assistance compresses the curve: weeks, not months." + }, + { + "span_id": "span-6b7d846eb9209199ba03f2ef", + "text": "02 \"I'm not a math genius.\" Python reads closer to English than to equations." + }, + { + "span_id": "span-a4c304db75b88c8e3139659c", + "text": "We teach the logic; you don't need the math." + }, + { + "span_id": "span-045207d0c79523e2ca9eb32c", + "text": "03 \"LLMs can code for me.\" They assist." + }, + { + "span_id": "span-3c581cccbc89ae20dd4c27b6", + "text": "You still need Python to build products that scale and to catch what the model gets wrong." + }, + { + "span_id": "span-4e4e06201853bbd3698c9228", + "text": "04 \"AI keeps changing.\" The course updates weekly, and we teach you how to keep teaching yourself as it evolves." + }, + { + "span_id": "span-6a1e38e8c8b4a49f42862d7d", + "text": "05 \"What if I get stuck?\" Course tutors on Discord, plus techniques for using AI itself to get unstuck." + }, + { + "span_id": "span-ba5d88f7a406bd4d19f5edef", + "text": "06 \"Am I too late?\" The field is still young." + }, + { + "span_id": "span-71a65a968c12253e3669f256", + "text": "Python remains the skill that compounds fastest with it." + }, + { + "span_id": "span-efbbbd87f8a7d25f53a058b2", + "text": "03 Student outcomes" + }, + { + "span_id": "span-f30becd46e3c2812f730efa9", + "text": "04 The live syllabus" + } + ] + }, + { + "chunk_id": "tai-53da553182d1a85f4d82ba1e", + "text": "Course curriculum\nSix parts, 38 lessons, a dozen projects, and a portfolio-ready capstone. The first lessons of every part are free to preview. 01 Introduction 4 lessons + 1 Course Introduction and Rationale Free preview 2 Course Structure Free preview 3 How LLMs can be used as Coding Assistants Free preview 4 Computer Science Fundamentals Free preview 02 Using LLMs 4 lessons + 1 Part 1 Overview: Using LLMs Free preview 2 Introduction to ChatGPT, Claude, and Gemini 3 Prompting 101 4 Quiz 03 Beginner code projects 8 lessons + 1 Part 2 Overview: Beginner Code Projects Free preview 2 Personal Budget Tracker Free preview 3 Healthy Habits Tracker 4 Quiz 5 Understanding Python Environments and Tools Free preview 6 Analyzing Coffee Shop Sales with Notebooks 7 Text Adventure Game 8 Quiz 04 Intermediate code projects 8 lessons + 1 Part 3 Overview: Intermediate Code Projects Free preview 2 GitHub and Code Versioning 3 Random Password Generator 4 Web Scraping 5 Quiz 6 Interactive Quiz App with Gradio 7 Personal Finance Dashboard with Charts 8 Quiz 05 Code projects using LLMs 8 lessons + 1 Part 4 Overview: Code Projects Using LLMs Free preview 2 How LLMs work 3 News Analyzer: Summarize, Sentiment, and Tags 4 Chat with Your Documents: A RAG Application 5 Create an API for Drafting Social Posts 6 Quiz 7 Future of AI-Driven Programming 8 The Future in Practice: Your Guide to AI Coding Tools 06 Next steps and capstone project 6 lessons + 1 Part 5 Overview: Next Steps and Capstone Project Free preview 2 Where to Go From Here: Paths to Explore 3 Capstone Project Description 4 Capstone Project DropBox 5 Share your success and showcase your new skills! 6 AI Tutor: Ask any question! Updated weekly with new lessons as the field evolves. $149 one-time · lifetime access 38 lessons, 12+", + "heading": "Course curriculum", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-f187b69c688c37a16ab46799", + "text": "Six parts, 38 lessons, a dozen projects, and a portfolio-ready capstone." + }, + { + "span_id": "span-e11a90cffe094e0e34f97b3b", + "text": "The first lessons of every part are free to preview." + }, + { + "span_id": "span-73bdf7d1c3c6b944ec9152da", + "text": "01 Introduction 4 lessons + 1 Course Introduction and Rationale Free preview" + }, + { + "span_id": "span-6822ba3966ba0d2d326bde1e", + "text": "2 Course Structure Free preview" + }, + { + "span_id": "span-4fff2db335f76afd6279b930", + "text": "3 How LLMs can be used as Coding Assistants Free preview" + }, + { + "span_id": "span-abb46dd58b96a8705e11083f", + "text": "4 Computer Science Fundamentals Free preview" + }, + { + "span_id": "span-516d5eb117a0c12217996285", + "text": "02 Using LLMs 4 lessons + 1 Part 1 Overview: Using LLMs Free preview" + }, + { + "span_id": "span-946b96ae4ac1b0855680070f", + "text": "2 Introduction to ChatGPT, Claude, and Gemini" + }, + { + "span_id": "span-a332664571a5eb2e341be181", + "text": "3 Prompting 101" + }, + { + "span_id": "span-221cf8748350df3fd9dcb715", + "text": "4 Quiz" + }, + { + "span_id": "span-835747b85128ee5354c9b5ce", + "text": "03 Beginner code projects 8 lessons + 1 Part 2 Overview: Beginner Code Projects Free preview" + }, + { + "span_id": "span-96b80baf428d3b24f71a5bf2", + "text": "2 Personal Budget Tracker Free preview" + }, + { + "span_id": "span-d75159c48a30321deb8d9e58", + "text": "3 Healthy Habits Tracker" + }, + { + "span_id": "span-86d2289013bc51023883adf3", + "text": "5 Understanding Python Environments and Tools Free preview" + }, + { + "span_id": "span-778c162c69aae3d9b757eea1", + "text": "6 Analyzing Coffee Shop Sales with Notebooks" + }, + { + "span_id": "span-2c42182a140c0e5d6c087705", + "text": "7 Text Adventure Game" + }, + { + "span_id": "span-916304884ae34f5374ddcc85", + "text": "8 Quiz" + }, + { + "span_id": "span-fb36ec31bf1c01b471e03670", + "text": "04 Intermediate code projects 8 lessons + 1 Part 3 Overview: Intermediate Code Projects Free preview" + }, + { + "span_id": "span-b3702c408e5b3548c66aac55", + "text": "2 GitHub and Code Versioning" + }, + { + "span_id": "span-f714f107e6d000e4f38ecf9d", + "text": "3 Random Password Generator" + }, + { + "span_id": "span-97c722368d6d0c8c8cb4c15d", + "text": "4 Web Scraping" + }, + { + "span_id": "span-779c0370feeb9c4dfa68097c", + "text": "5 Quiz" + }, + { + "span_id": "span-05816275024b1db49261ba47", + "text": "6 Interactive Quiz App with Gradio" + }, + { + "span_id": "span-b164faa423159043811b22f3", + "text": "7 Personal Finance Dashboard with Charts" + }, + { + "span_id": "span-d81a6ee543f9f468ec0efe59", + "text": "05 Code projects using LLMs 8 lessons + 1 Part 4 Overview: Code Projects Using LLMs Free preview" + }, + { + "span_id": "span-1c371f768ffac6f013888ddc", + "text": "2 How LLMs work" + }, + { + "span_id": "span-6a777497f1d9585db292aa29", + "text": "3 News Analyzer: Summarize, Sentiment, and Tags" + }, + { + "span_id": "span-d83fbde6e34bea522ebb828c", + "text": "4 Chat with Your Documents: A RAG Application" + }, + { + "span_id": "span-49734e71ffeb619b47075f4e", + "text": "5 Create an API for Drafting Social Posts" + }, + { + "span_id": "span-871d793d9001b2e9eed3584a", + "text": "6 Quiz" + }, + { + "span_id": "span-81233665283a953f1a276222", + "text": "7 Future of AI-Driven Programming" + }, + { + "span_id": "span-abd69d452981ed19d31597c8", + "text": "8 The Future in Practice: Your Guide to AI Coding Tools" + }, + { + "span_id": "span-6dc778f4c138450f560927ae", + "text": "06 Next steps and capstone project 6 lessons + 1 Part 5 Overview: Next Steps and Capstone Project Free preview" + }, + { + "span_id": "span-a600250f21720538fc882310", + "text": "2 Where to Go From Here: Paths to Explore" + }, + { + "span_id": "span-846dd778968422eec90ee0d0", + "text": "3 Capstone Project Description" + }, + { + "span_id": "span-bb6db649bd0bc35937e833a2", + "text": "4 Capstone Project DropBox" + }, + { + "span_id": "span-1c0724be6770c6e590971c2f", + "text": "5 Share your success and showcase your new skills!" + }, + { + "span_id": "span-3e1009cd7751d5d7ab9bbc2c", + "text": "6 AI Tutor: Ask any question!" + }, + { + "span_id": "span-b5d495fc1286cbcd787b96de", + "text": "Updated weekly with new lessons as the field evolves." + }, + { + "span_id": "span-393ad17faa5d1a8d4ef28d63", + "text": "$149 one-time · lifetime access" + } + ] + }, + { + "chunk_id": "tai-1fd67638401c022af8ecc0d0", + "text": "Course curriculum\nhands-on projects Portfolio-ready capstone project Lifetime access, updated weekly Certification on completion Discord tutors and AI tutor support Unlock lifetime access 30-day money-back guarantee, no questions asked 05 Start from zero\n\nWho is this course for?\nComplete beginners, aspiring AI developers, and professionals who want to lead AI projects. No prior coding, no software background: just curiosity, a computer, and internet access. From black box to toolbox\n\nGo beyond prompting\nMost people just use AI. With Python you start customising, fine-tuning, and building AI solutions that actually fit your needs. The hiring signal\n\nOpen the door to AI roles\nPython is the language of AI. Whether you want to automate tasks, analyse data, or fine-tune models, it's the skill companies look for first. Build, don't wait\n\nTools that work for you\nA chatbot, a research assistant, an automation: with Python you build what you need instead of relying on whatever tool happens to exist. + Next step after this course: Full Stack AI Engineering , where these Python skills become production LLM systems. 06 Questions", + "heading": "Tools that work for you", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-443097d6d897463b205fd65b", + "text": "Unlock lifetime access 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-b3c1049c238cac7888615a00", + "text": "05 Start from zero" + }, + { + "span_id": "span-394c5422387cda68d98cb00a", + "text": "Complete beginners, aspiring AI developers, and professionals who want to lead AI projects." + }, + { + "span_id": "span-c8e8e9d15cc6d67310f3f924", + "text": "No prior coding, no software background: just curiosity, a computer, and internet access." + }, + { + "span_id": "span-efc73cd3fa1e68712ca1f088", + "text": "From black box to toolbox" + }, + { + "span_id": "span-9f1883534280995db872a6d9", + "text": "Most people just use AI." + }, + { + "span_id": "span-04a694ed7601e99e87a426a6", + "text": "With Python you start customising, fine-tuning, and building AI solutions that actually fit your needs." + }, + { + "span_id": "span-7374c7bfc5b83d4001cc39cb", + "text": "The hiring signal" + }, + { + "span_id": "span-695eb4647c1920485118331c", + "text": "Python is the language of AI." + }, + { + "span_id": "span-a241440a8b5d46ec02d499f4", + "text": "Whether you want to automate tasks, analyse data, or fine-tune models, it's the skill companies look for first." + }, + { + "span_id": "span-64a25539f3554d54d6617146", + "text": "Build, don't wait" + }, + { + "span_id": "span-3e428671480991f3a8dcfc0e", + "text": "A chatbot, a research assistant, an automation: with Python you build what you need instead of relying on whatever tool happens to exist." + }, + { + "span_id": "span-680b684f9ede261abc757b5c", + "text": "+ Next step after this course: Full Stack AI Engineering , where these Python skills become production LLM systems." + }, + { + "span_id": "span-8708bca7364ee8d54cb66113", + "text": "06 Questions" + } + ] + }, + { + "chunk_id": "tai-1c981ab255ae3fb93568eb64", + "text": "Frequently asked questions\nDo I need prior experience with coding or AI? + No. The course teaches Python from scratch and assumes no software background. Will I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What projects will I work on? + A dozen builds from a personal budget tracker to web scraping, a Gradio quiz app, a news analyzer, and a chat-with-your-documents RAG application, ending in a capstone project of your own. Is it still up to date? + The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. What is your refund policy? + If, after completing the first section, you feel the course isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries. Lifetime access · $149\n\nLearn Python the LLM-native way\nCreate with LLMs and transform your work with code, starting from zero. Start learning smarter See the full pathway", + "heading": "Frequently asked questions", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-8d1a1780f23d791a7a91bb3c", + "text": "The course teaches Python from scratch and assumes no software background." + }, + { + "span_id": "span-2385a2300032dc2abec0e5fd", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-bd328da226a1245d46516a06", + "text": "A dozen builds from a personal budget tracker to web scraping, a Gradio quiz app, a news analyzer, and a chat-with-your-documents RAG application, ending in a capstone project of your own." + }, + { + "span_id": "span-344d86b23c65e8229fb3ed45", + "text": "The course launched in 2025 and is updated every week with new lessons, tools, and techniques as the AI landscape evolves." + }, + { + "span_id": "span-6c3aa357aabd810f1ba1f02f", + "text": "Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing." + }, + { + "span_id": "span-dd856b290542386c87362ae6", + "text": "Contact us to claim." + }, + { + "span_id": "span-752c1537ce793208a68b8857", + "text": "If, after completing the first section, you feel the course isn't right for you, we'll provide a 100% refund within 30 days." + }, + { + "span_id": "span-cb6a869a22952d873184b1be", + "text": "No hassle, no worries." + }, + { + "span_id": "span-2ca01c6b55ad4b879cc56496", + "text": "Lifetime access · $149" + }, + { + "span_id": "span-dc6fc6726ae6e21512f8c4af", + "text": "Create with LLMs and transform your work with code, starting from zero." + }, + { + "span_id": "span-2c2e66e421f4e2357b2ed220", + "text": "Start learning smarter See the full pathway" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "a61e6ddc73ec3aa8de1162687ba7619c3b81fbae99a8b7c1f4d2deed31874745", + "content_hash": "a61e6ddc73ec3aa8de1162687ba7619c3b81fbae99a8b7c1f4d2deed31874745", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/python-for-ai-engineering/" + ], + "evidence_hash": "d7c8d6e478e56026ae9325ec88655a83b521ffaa223298f277f0a283160b3c06" + }, + { + "discovered_url": "https://towardsai.com/academy/ai-for-work/", + "url": "https://towardsai.com/academy/ai-for-work/", + "canonical_url": "https://towardsai.com/academy/ai-for-work/", + "host": "towardsai.com", + "path": "/academy/ai-for-work", + "kind": "course", + "offer_id": "ai-for-work", + "entity_id": "offer:ai-for-work", + "title": "Master AI for Work: No-Code AI for Professionals | Towards AI Academy", + "review_title": "AI for Work", + "reviewed_summary": "Practical no-code AI training for professionals.", + "meta_description": "Get real benefits from LLMs at work without writing code: prompting, research, analysis, agents, and role-specific modules for sales, marketing, finance, HR, and leadership. $399 lifetime access.", + "headings": [ + "Master AI for work", + "Drive AI adoption in your team, company, and industry", + "Use AI tools for real work", + "Kickstart custom AI projects", + "Case studies that saved millions", + "Who is this course for?", + "Course curriculum", + "What you'll learn, and actually use", + "The right tool, every time", + "Four assistants in one", + "Consistently great outputs", + "Analysis without code", + "Workflows and agents", + "Smart adoption decisions", + "Frequently asked questions", + "Become the AI power user on your team" + ], + "text": "No code required · Self-paced\n\nMaster AI for work\nGet real benefits from LLMs on valuable tasks, taught by experts who build AI for a living. Prompting, research, analysis, agents, and role-specific modules, all without writing code. See the modules Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught 98 lessons across core and role modules 10 role-specific tracks, sales to leadership 0 lines of code required Pay once 3 monthly payments $495 $399 one-time Save 19% Pay once and save $96 versus $495 on the 3-month plan. $165 /month for 3 months $495 in total, with the same lifetime access. Learn Prompting, research, analysis, and agents, all without code Format 98 lessons plus 10 role-specific tracks, sales to leadership Outcome Certification and workflows you can use at work the same day Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 The AI-first mindset\n\nDrive AI adoption in your team, company, and industry\nAI is no longer a nice-to-have; it's a boardroom mandate. This course makes you the person who knows what to do about it. Beyond basic prompting\n\nUse AI tools for real work\nWork with ChatGPT, Gemini, Claude, and Perplexity to automate your tasks, and master prompting, context building, and reasoning techniques that produce high-quality output every time. No technical know-how needed\n\nKickstart custom AI projects\nMaster no-code tools like NotebookLM, Deep Research, CustomGPTs, agents, and Artifacts to automate repetitive work, personalise customer experiences, and lift your team's productivity. Learn from what worked\n\nCase studies that saved millions\nActionable insights from real AI implementations, including how companies like Klarna cut costs at scale, plus the strategy, implementation, and governance lens to replicate it in your context. 01 Our AI-first alumni combine their own expertise with AI to deliver higher quality and quantity of work relative to colleagues. 02 They find where AI tools add the most value in everyday workflows, and get promoted faster by delivering results quicker. 02 Every professional role\n\nWho is this course for?\nAnyone ready to become an AI power user at work, with dedicated role modules so the examples match your job. 01 Any professional ready to kickstart AI at work. Role modules cover sales, marketing, investment research and banking, HR and recruitment, product management, software development, customer service, consulting, and leadership. 02 Managers, team leaders, and executives guiding smart adoption, with a dedicated module on leading AI-first initiatives internally. 03 Consultants and strategists advising clients in the AI era. 04 Entrepreneurs and business owners wanting to scale with fewer roadblocks. 03 The live syllabus\n\nCourse curriculum\nA hands-on core plus optional role-specific modules, 98 lessons in all. The first lessons are free to preview. 00 Introduction 3 lessons + 1 Introduction Free preview 2 Course Admin and Syllabus Free preview 3 Top Tips Cheat Sheet Free preview 01 Getting started with generative AI 9 lessons + 1 Introduction to Module 1 Free preview 2 Prompting 101 Free preview 3 A Brief Introduction to AI and LLMs Free preview 4 Alien Intelligence Leads to Unintuitive Mistakes 5 Introduction to ChatGPT, Claude and Gemini 6 Security and Privacy Risks When Using LLMs in Business 7 Providing \"Context\" to LLMs for Better Results, plus RAG Chatbots 8 Claude Cowork Part 1: Your AI Teammate That Actually Does the Work 9 Claude Cowork Part 2: Claude in Chrome, Excel and PPT + 10 Hands-on Use Cases 02 Core LLM use cases with hands-on examples 23 lessons + 1 AI as a personal tutor, research assistant, and writing partner 2 NotebookLM, Perplexity, and deep research workflows 3 Summarisation, parsing, and data extraction 4 Multimodal AI: images, graphics, and infographics with Artifacts 5 Choosing your LLM: benchmarks, leaderboards, and reasoning models 6 Data analysis and visualisation without code 7 AI agents and deep research across OpenAI, Gemini, and Perplexity 8 Case study: organising a major conference end to end 9 Robust prompts, Custom GPTs, evaluation, and fact verification 03 More models, agents, and tools 7 lessons + 1 An overview of leading LLMs: model families, classes, and comparison sheet 2 No-code platforms for building custom AI agents 3 Hands-on agent examples: Integrail, CrewAI, and Copilot Studio 4 Staying ahead of AI developments 04 Role-specific modules (optional) 10 tracks + 1 Sales: email sequences, B2B cycles, and case studies 2 Marketing: product launches, customer engagement, campaign case studies 3 Investment research and finance: theses, rapid response, trading use cases 4 HR and recruitment: wellness programs and case studies 5 Product management and innovation: the full lifecycle with AI 6 Software and \"vibe coding\": Copilot, Cursor, and coding methods 7 Accounting and finance: QBR preparation and reporting 8 Customer service and support: multi-step resolution workflows 9 Professional consulting: multi-step client workflows 10 Management and leadership: team efficiency and adoption 05 AI strategy for leaders 11 lessons + 1 The generative AI landscape and enterprise adoption 2 Build vs buy: custom and off-the-shelf solutions 3 Defining your own AI tool by industry, niche, function, and users 4 Customising LLMs: prompting, RAG, tools, and fine-tuning 5 Ethical and responsible AI use 6 Managing custom LLM development, surveys, and customer case studies Updated weekly with new lessons as the field evolves. Pay once 3 monthly payments $495 $399 one-time Save 19% Pay once and save $96 versus $495 on the 3-month plan. $165 /month for 3 months $495 in total, with the same lifetime access. Learn Prompting, research, analysis, and agents, all without code Format 98 lessons plus 10 role-specific tracks, sales to leadership Outcome Certification and workflows you can use at work the same day Lifetime access, updated weekly Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 04 Practical from day one\n\nWhat you'll learn, and actually use\n01\n\nThe right tool, every time\nChatGPT, Claude, Gemini, NotebookLM, and Perplexity: know which to use, when, and why. 02\n\nFour assistants in one\nUse AI as your personal tutor, research assistant, brainstorming partner, and design intern. 03\n\nConsistently great outputs\nMaster prompting, context, and reasoning so quality stops depending on luck. 04\n\nAnalysis without code\nAnalyse data and build visuals directly in AI tools, no spreadsheet formulas required. 05\n\nWorkflows and agents\nCreate AI workflows and agents that save time and scale your results. 06\n\nSmart adoption decisions\nMake sound calls on tools, privacy, governance, and team adoption. 05 Questions\n\nFrequently asked questions\nDo I need any technical background? + No. The course is designed for non-technical professionals and focuses on practical, no-code applications you can implement immediately. How long does it take to complete? + Self-paced; the average completion time for the core course is around 20 hours, with optional role-specific modules on top. Will I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What real-world skills will I gain? + How to implement AI tools like ChatGPT in your daily work, customise AI solutions for your company, and lead AI-driven projects that improve efficiency and results. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. Can I take this course within my company? + Yes. We offer course bundles and custom training tailored for companies, including live formats. See enterprise enablement or get in touch . Lifetime access · $399\n\nBecome the AI power user on your team\nBoost efficiency, enhance your team's capabilities, and spot the AI opportunities in your sector before your competitors do. Enroll now See the full pathway", + "links": [ + { + "text": "Unlock lifetime access", + "url": "https://academy.towardsai.net/cart/add_product/3147325?price_id=4025392" + }, + { + "text": "Start the 3-month plan", + "url": "https://academy.towardsai.net/enroll/3147325?price_id=4431469" + }, + { + "text": "Introduction Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "Course Admin and Syllabus Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "Top Tips Cheat Sheet Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "Introduction to Module 1 Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "Prompting 101 Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "A Brief Introduction to AI and LLMs Free preview", + "url": "https://academy.towardsai.net/enroll/3147325?et=free_trial" + }, + { + "text": "Contact us", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "enterprise enablement", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "get in touch", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "Enroll now", + "url": "https://academy.towardsai.net/cart/add_product/3147325?price_id=4025392" + }, + { + "text": "See the full pathway", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-0a5d164aaa5bb0cf887b0ddb", + "text": "No code required · Self-paced\n\nMaster AI for work\nGet real benefits from LLMs on valuable tasks, taught by experts who build AI for a living. Prompting, research, analysis, agents, and role-specific modules, all without writing code. See the modules Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught 98 lessons across core and role modules 10 role-specific tracks, sales to leadership 0 lines of code required Pay once 3 monthly payments $495 $399 one-time Save 19% Pay once and save $96 versus $495 on the 3-month plan. $165 /month for 3 months $495 in total, with the same lifetime access. Learn Prompting, research, analysis, and agents, all without code Format 98 lessons plus 10 role-specific tracks, sales to leadership Outcome Certification and workflows you can use at work the same day Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 01 The AI-first mindset\n\nDrive AI adoption in your team, company, and industry\nAI is no longer a nice-to-have; it's a boardroom mandate. This course makes you the person who knows what to do about it. Beyond basic prompting\n\nUse AI tools for real work\nWork with ChatGPT, Gemini, Claude, and Perplexity to automate your tasks, and master prompting, context building, and reasoning techniques that produce high-quality output every time. No technical know-how needed\n\nKickstart custom AI projects\nMaster no-code tools like NotebookLM, Deep Research, CustomGPTs, agents, and Artifacts to automate repetitive work, personalise customer experiences, and lift your team's productivity. Learn from what worked", + "heading": "Master AI for work", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-6426af523b373c98f4a2ae64", + "text": "No code required · Self-paced" + }, + { + "span_id": "span-33b91b44e71b3261979d4846", + "text": "Get real benefits from LLMs on valuable tasks, taught by experts who build AI for a living." + }, + { + "span_id": "span-30110b820ababdb02044eb55", + "text": "Prompting, research, analysis, agents, and role-specific modules, all without writing code." + }, + { + "span_id": "span-d3c5fa314694cf58654f4217", + "text": "See the modules" + }, + { + "span_id": "span-f1ccc432dfd6410aaad99946", + "text": "Taught by Louis-François Bouchard and the Towards AI team Co-founder & CTO, Towards AI · 500,000+ AI practitioners taught" + }, + { + "span_id": "span-73454e16a7491781691c9f5c", + "text": "98 lessons across core and role modules" + }, + { + "span_id": "span-49883b56a78c3d283a73f364", + "text": "10 role-specific tracks, sales to leadership" + }, + { + "span_id": "span-40ce791e82cccab5fa6b3fae", + "text": "0 lines of code required" + }, + { + "span_id": "span-9a7b38f23e887e356df799a0", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-1fbb3f838bb1f75dc3fc4401", + "text": "$495 $399 one-time Save 19%" + }, + { + "span_id": "span-27df7fb8a84ae64c84d237e7", + "text": "Pay once and save $96 versus $495 on the 3-month plan." + }, + { + "span_id": "span-3ca238e8da196e7d3e1b0b4f", + "text": "$165 /month for 3 months" + }, + { + "span_id": "span-2d0059e45d385085876ffbd9", + "text": "$495 in total, with the same lifetime access." + }, + { + "span_id": "span-b3cdb4df793c42163149d4a0", + "text": "Prompting, research, analysis, and agents, all without code" + }, + { + "span_id": "span-8e48445820c351edf0048b22", + "text": "98 lessons plus 10 role-specific tracks, sales to leadership" + }, + { + "span_id": "span-f262499ea37a7da49c740102", + "text": "Certification and workflows you can use at work the same day" + }, + { + "span_id": "span-81139f05adf3234b069b65ab", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-9425dd813df9681f3992e67a", + "text": "01 The AI-first mindset" + }, + { + "span_id": "span-e371b59c376d458033784892", + "text": "AI is no longer a nice-to-have; it's a boardroom mandate." + }, + { + "span_id": "span-6850d7204dec5454cd6e2a69", + "text": "This course makes you the person who knows what to do about it." + }, + { + "span_id": "span-49aa9e7f4663dfce5d532c36", + "text": "Beyond basic prompting" + }, + { + "span_id": "span-857988b7f64d524808b839ea", + "text": "Work with ChatGPT, Gemini, Claude, and Perplexity to automate your tasks, and master prompting, context building, and reasoning techniques that produce high-quality output every time." + }, + { + "span_id": "span-132a871ed4402349a4524faf", + "text": "No technical know-how needed" + }, + { + "span_id": "span-5b31dbe2c75abcb722a15e4b", + "text": "Master no-code tools like NotebookLM, Deep Research, CustomGPTs, agents, and Artifacts to automate repetitive work, personalise customer experiences, and lift your team's productivity." + }, + { + "span_id": "span-3f47cae351dd90e7755b4e05", + "text": "Learn from what worked" + } + ] + }, + { + "chunk_id": "tai-a083565eee0081072e561bd9", + "text": "Case studies that saved millions\nActionable insights from real AI implementations, including how companies like Klarna cut costs at scale, plus the strategy, implementation, and governance lens to replicate it in your context. 01 Our AI-first alumni combine their own expertise with AI to deliver higher quality and quantity of work relative to colleagues. 02 They find where AI tools add the most value in everyday workflows, and get promoted faster by delivering results quicker. 02 Every professional role\n\nWho is this course for?\nAnyone ready to become an AI power user at work, with dedicated role modules so the examples match your job. 01 Any professional ready to kickstart AI at work. Role modules cover sales, marketing, investment research and banking, HR and recruitment, product management, software development, customer service, consulting, and leadership. 02 Managers, team leaders, and executives guiding smart adoption, with a dedicated module on leading AI-first initiatives internally. 03 Consultants and strategists advising clients in the AI era. 04 Entrepreneurs and business owners wanting to scale with fewer roadblocks. 03 The live syllabus", + "heading": "Who is this course for?", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-cee4b97c150517eb59b38648", + "text": "Actionable insights from real AI implementations, including how companies like Klarna cut costs at scale, plus the strategy, implementation, and governance lens to replicate it in your context." + }, + { + "span_id": "span-e34e3b7fddc2037c1fa81409", + "text": "01 Our AI-first alumni combine their own expertise with AI to deliver higher quality and quantity of work relative to colleagues." + }, + { + "span_id": "span-86d4246d10162332bf959a45", + "text": "02 They find where AI tools add the most value in everyday workflows, and get promoted faster by delivering results quicker." + }, + { + "span_id": "span-6d1bdc3d6c9e69a82619df5a", + "text": "02 Every professional role" + }, + { + "span_id": "span-275ff49c63de38029c15297d", + "text": "Anyone ready to become an AI power user at work, with dedicated role modules so the examples match your job." + }, + { + "span_id": "span-9c8157928e67d1304e7be489", + "text": "01 Any professional ready to kickstart AI at work." + }, + { + "span_id": "span-05406f7eb67def840f09b7a5", + "text": "Role modules cover sales, marketing, investment research and banking, HR and recruitment, product management, software development, customer service, consulting, and leadership." + }, + { + "span_id": "span-10471491f8201b7076102b8b", + "text": "02 Managers, team leaders, and executives guiding smart adoption, with a dedicated module on leading AI-first initiatives internally." + }, + { + "span_id": "span-05261ae8cdb12cb168358c82", + "text": "03 Consultants and strategists advising clients in the AI era." + }, + { + "span_id": "span-74afb9dc41c539ac272ede51", + "text": "04 Entrepreneurs and business owners wanting to scale with fewer roadblocks." + }, + { + "span_id": "span-3afd5e36f8d2dd1b61c8a858", + "text": "03 The live syllabus" + } + ] + }, + { + "chunk_id": "tai-0fb166183a0e26a78a4e3164", + "text": "Course curriculum\nA hands-on core plus optional role-specific modules, 98 lessons in all. The first lessons are free to preview. 00 Introduction 3 lessons + 1 Introduction Free preview 2 Course Admin and Syllabus Free preview 3 Top Tips Cheat Sheet Free preview 01 Getting started with generative AI 9 lessons + 1 Introduction to Module 1 Free preview 2 Prompting 101 Free preview 3 A Brief Introduction to AI and LLMs Free preview 4 Alien Intelligence Leads to Unintuitive Mistakes 5 Introduction to ChatGPT, Claude and Gemini 6 Security and Privacy Risks When Using LLMs in Business 7 Providing \"Context\" to LLMs for Better Results, plus RAG Chatbots 8 Claude Cowork Part 1: Your AI Teammate That Actually Does the Work 9 Claude Cowork Part 2: Claude in Chrome, Excel and PPT + 10 Hands-on Use Cases 02 Core LLM use cases with hands-on examples 23 lessons + 1 AI as a personal tutor, research assistant, and writing partner 2 NotebookLM, Perplexity, and deep research workflows 3 Summarisation, parsing, and data extraction 4 Multimodal AI: images, graphics, and infographics with Artifacts 5 Choosing your LLM: benchmarks, leaderboards, and reasoning models 6 Data analysis and visualisation without code 7 AI agents and deep research across OpenAI, Gemini, and Perplexity 8 Case study: organising a major conference end to end 9 Robust prompts, Custom GPTs, evaluation, and fact verification 03 More models, agents, and tools 7 lessons + 1 An overview of leading LLMs: model families, classes, and comparison sheet 2 No-code platforms for building custom AI agents 3 Hands-on agent examples: Integrail, CrewAI, and Copilot Studio 4 Staying ahead of AI developments 04 Role-specific modules (optional) 10 tracks + 1 Sales: email sequences, B2B cycles, and case studies 2 Marketing: product", + "heading": "Course curriculum", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-0de0ecac18142c4f73be28b4", + "text": "A hands-on core plus optional role-specific modules, 98 lessons in all." + }, + { + "span_id": "span-efdc1a8c29a29d03ba0ec503", + "text": "The first lessons are free to preview." + }, + { + "span_id": "span-560130ee5d1c4591bca3d8c5", + "text": "00 Introduction 3 lessons + 1 Introduction Free preview" + }, + { + "span_id": "span-109ee5671b300dae00c8d289", + "text": "2 Course Admin and Syllabus Free preview" + }, + { + "span_id": "span-7a289d25680f5295b9b28958", + "text": "3 Top Tips Cheat Sheet Free preview" + }, + { + "span_id": "span-26edb8ba6936e849c0b9b43f", + "text": "01 Getting started with generative AI 9 lessons + 1 Introduction to Module 1 Free preview" + }, + { + "span_id": "span-432d10fa59630246392e37a4", + "text": "2 Prompting 101 Free preview" + }, + { + "span_id": "span-b13ac4ac061f0c96bce00938", + "text": "3 A Brief Introduction to AI and LLMs Free preview" + }, + { + "span_id": "span-10eeddf2ab320a731f0edde8", + "text": "4 Alien Intelligence Leads to Unintuitive Mistakes" + }, + { + "span_id": "span-45e128d5496adea664f24028", + "text": "5 Introduction to ChatGPT, Claude and Gemini" + }, + { + "span_id": "span-1c3e83cb172fa4bcfc196995", + "text": "6 Security and Privacy Risks When Using LLMs in Business" + }, + { + "span_id": "span-cd8d9047d42664ba1b32e7f8", + "text": "7 Providing \"Context\" to LLMs for Better Results, plus RAG Chatbots" + }, + { + "span_id": "span-598f4d72b67545ffa55d82ae", + "text": "8 Claude Cowork Part 1: Your AI Teammate That Actually Does the Work" + }, + { + "span_id": "span-4961810f72b6d80be49ddf9f", + "text": "9 Claude Cowork Part 2: Claude in Chrome, Excel and PPT + 10 Hands-on Use Cases" + }, + { + "span_id": "span-f6175c9d55966801fcabfed6", + "text": "02 Core LLM use cases with hands-on examples 23 lessons + 1 AI as a personal tutor, research assistant, and writing partner" + }, + { + "span_id": "span-9315cd5698d6a1f1ccaec5ad", + "text": "2 NotebookLM, Perplexity, and deep research workflows" + }, + { + "span_id": "span-a429cbac7730242dc49042d5", + "text": "3 Summarisation, parsing, and data extraction" + }, + { + "span_id": "span-5d812f0fdb6a4dcdde244a72", + "text": "4 Multimodal AI: images, graphics, and infographics with Artifacts" + }, + { + "span_id": "span-0c1ee2ded0511563e3dc3c48", + "text": "5 Choosing your LLM: benchmarks, leaderboards, and reasoning models" + }, + { + "span_id": "span-e3ac99d967972e717148efec", + "text": "6 Data analysis and visualisation without code" + }, + { + "span_id": "span-0a5dd311b933a9323bf0eb90", + "text": "7 AI agents and deep research across OpenAI, Gemini, and Perplexity" + }, + { + "span_id": "span-54d8899ad7515800123d4005", + "text": "8 Case study: organising a major conference end to end" + }, + { + "span_id": "span-e494e17dfaaf3e5dd17ceb72", + "text": "9 Robust prompts, Custom GPTs, evaluation, and fact verification" + }, + { + "span_id": "span-e624e81c5c6b77a099fe6cb9", + "text": "03 More models, agents, and tools 7 lessons + 1 An overview of leading LLMs: model families, classes, and comparison sheet" + }, + { + "span_id": "span-2100b63a6a4515d0e1e2fbf9", + "text": "2 No-code platforms for building custom AI agents" + }, + { + "span_id": "span-4424839e346aed23cc543c7e", + "text": "3 Hands-on agent examples: Integrail, CrewAI, and Copilot Studio" + }, + { + "span_id": "span-bccbbd867a24de504572e0c6", + "text": "4 Staying ahead of AI developments" + }, + { + "span_id": "span-c77fe823fbb123dbfecf471e", + "text": "04 Role-specific modules (optional) 10 tracks + 1 Sales: email sequences, B2B cycles, and case studies" + } + ] + }, + { + "chunk_id": "tai-5b6f50ac1a9872dc0f03fc8b", + "text": "Course curriculum\nlaunches, customer engagement, campaign case studies 3 Investment research and finance: theses, rapid response, trading use cases 4 HR and recruitment: wellness programs and case studies 5 Product management and innovation: the full lifecycle with AI 6 Software and \"vibe coding\": Copilot, Cursor, and coding methods 7 Accounting and finance: QBR preparation and reporting 8 Customer service and support: multi-step resolution workflows 9 Professional consulting: multi-step client workflows 10 Management and leadership: team efficiency and adoption 05 AI strategy for leaders 11 lessons + 1 The generative AI landscape and enterprise adoption 2 Build vs buy: custom and off-the-shelf solutions 3 Defining your own AI tool by industry, niche, function, and users 4 Customising LLMs: prompting, RAG, tools, and fine-tuning 5 Ethical and responsible AI use 6 Managing custom LLM development, surveys, and customer case studies Updated weekly with new lessons as the field evolves. Pay once 3 monthly payments $495 $399 one-time Save 19% Pay once and save $96 versus $495 on the 3-month plan. $165 /month for 3 months $495 in total, with the same lifetime access. Learn Prompting, research, analysis, and agents, all without code Format 98 lessons plus 10 role-specific tracks, sales to leadership Outcome Certification and workflows you can use at work the same day Lifetime access, updated weekly Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked 04 Practical from day one\n\nWhat you'll learn, and actually use\n01", + "heading": "Course curriculum", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-03a1b82dadffed9b21dd4d88", + "text": "3 Investment research and finance: theses, rapid response, trading use cases" + }, + { + "span_id": "span-f3803ac27980fb25e737e059", + "text": "4 HR and recruitment: wellness programs and case studies" + }, + { + "span_id": "span-2fdedb718f246577a651e947", + "text": "5 Product management and innovation: the full lifecycle with AI" + }, + { + "span_id": "span-a52ba8e49a07ed3542089b4c", + "text": "6 Software and \"vibe coding\": Copilot, Cursor, and coding methods" + }, + { + "span_id": "span-5ebe6af3e1cbd7617e7bdbd5", + "text": "7 Accounting and finance: QBR preparation and reporting" + }, + { + "span_id": "span-2ab3adf5f726436787fbbd05", + "text": "8 Customer service and support: multi-step resolution workflows" + }, + { + "span_id": "span-c1bea52e512e31369e123c2c", + "text": "9 Professional consulting: multi-step client workflows" + }, + { + "span_id": "span-0b9801fdccb45e60bb97df49", + "text": "10 Management and leadership: team efficiency and adoption" + }, + { + "span_id": "span-c665c65a9816589e68c6db20", + "text": "05 AI strategy for leaders 11 lessons + 1 The generative AI landscape and enterprise adoption" + }, + { + "span_id": "span-716fd395453c6bad7df2de31", + "text": "2 Build vs buy: custom and off-the-shelf solutions" + }, + { + "span_id": "span-aaea7aeafd4c958533480d1d", + "text": "3 Defining your own AI tool by industry, niche, function, and users" + }, + { + "span_id": "span-494da13b896185f57b9bfcb5", + "text": "4 Customising LLMs: prompting, RAG, tools, and fine-tuning" + }, + { + "span_id": "span-070b3f86e1963b324a398dd9", + "text": "5 Ethical and responsible AI use" + }, + { + "span_id": "span-d7283b936e7025350927b26e", + "text": "6 Managing custom LLM development, surveys, and customer case studies" + }, + { + "span_id": "span-b61098e9bfd480ba5a784de1", + "text": "Updated weekly with new lessons as the field evolves." + }, + { + "span_id": "span-b59372d14d059b8ecb2c4de5", + "text": "Pay once 3 monthly payments" + }, + { + "span_id": "span-b0e561dbd503f533dfc8bbb8", + "text": "$495 $399 one-time Save 19%" + }, + { + "span_id": "span-d1135f66644c20ab8be1e278", + "text": "Pay once and save $96 versus $495 on the 3-month plan." + }, + { + "span_id": "span-c27287cf14a15f82462ecf97", + "text": "$165 /month for 3 months" + }, + { + "span_id": "span-335b6099c1eb1a4f7ddc863b", + "text": "$495 in total, with the same lifetime access." + }, + { + "span_id": "span-2cc749ae0ef57eab8701372f", + "text": "Prompting, research, analysis, and agents, all without code" + }, + { + "span_id": "span-d1e1f6199d054bc47b69d909", + "text": "98 lessons plus 10 role-specific tracks, sales to leadership" + }, + { + "span_id": "span-5b6b1f5dc407bcc4c2e6491b", + "text": "Certification and workflows you can use at work the same day" + }, + { + "span_id": "span-902a55f6f3065356c9b2bc41", + "text": "Lifetime access, updated weekly" + }, + { + "span_id": "span-b3930146b091362f2d06df0f", + "text": "Unlock lifetime access Start the 3-month plan 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-957eb0639808cd864c7f9b0c", + "text": "04 Practical from day one" + }, + { + "span_id": "span-3a38590fde986854546c95dd", + "text": "What you'll learn, and actually use" + } + ] + }, + { + "chunk_id": "tai-093039fdcf615416a7099d62", + "text": "The right tool, every time\nChatGPT, Claude, Gemini, NotebookLM, and Perplexity: know which to use, when, and why. 02\n\nFour assistants in one\nUse AI as your personal tutor, research assistant, brainstorming partner, and design intern. 03\n\nConsistently great outputs\nMaster prompting, context, and reasoning so quality stops depending on luck. 04\n\nAnalysis without code\nAnalyse data and build visuals directly in AI tools, no spreadsheet formulas required. 05\n\nWorkflows and agents\nCreate AI workflows and agents that save time and scale your results. 06\n\nSmart adoption decisions\nMake sound calls on tools, privacy, governance, and team adoption. 05 Questions\n\nFrequently asked questions\nDo I need any technical background? + No. The course is designed for non-technical professionals and focuses on practical, no-code applications you can implement immediately. How long does it take to complete? + Self-paced; the average completion time for the core course is around 20 hours, with optional role-specific modules on top. Will I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. What real-world skills will I gain? + How to implement AI tools like ChatGPT in your daily work, customise AI solutions for your company, and lead AI-driven projects that improve efficiency and results. Can I get a discount? + Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing. Contact us to claim. Can I take this course within my company? + Yes. We offer course bundles and custom training tailored for companies, including live formats. See enterprise enablement or get in touch . Lifetime access · $399", + "heading": "Frequently asked questions", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-78163f7cecf12d3c548e4beb", + "text": "ChatGPT, Claude, Gemini, NotebookLM, and Perplexity: know which to use, when, and why." + }, + { + "span_id": "span-5cb8a83589123f730942a2eb", + "text": "Use AI as your personal tutor, research assistant, brainstorming partner, and design intern." + }, + { + "span_id": "span-3c06370cf1d9d00c55935652", + "text": "Master prompting, context, and reasoning so quality stops depending on luck." + }, + { + "span_id": "span-9bae7423c1d97ba592068d63", + "text": "Analyse data and build visuals directly in AI tools, no spreadsheet formulas required." + }, + { + "span_id": "span-bde831d883f61c8a7ba4b7bf", + "text": "Create AI workflows and agents that save time and scale your results." + }, + { + "span_id": "span-5a2952e75e080fdfce62dc69", + "text": "Make sound calls on tools, privacy, governance, and team adoption." + }, + { + "span_id": "span-c104c0d8bd20a8df36279ae4", + "text": "05 Questions" + }, + { + "span_id": "span-cf98b8f456a7cc4a74a3ddb3", + "text": "The course is designed for non-technical professionals and focuses on practical, no-code applications you can implement immediately." + }, + { + "span_id": "span-c58badd4b57cf26dd2b60358", + "text": "Self-paced; the average completion time for the core course is around 20 hours, with optional role-specific modules on top." + }, + { + "span_id": "span-0844be6311e3fcce3f353528", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-db808ddff3412735859a4981", + "text": "How to implement AI tools like ChatGPT in your daily work, customise AI solutions for your company, and lead AI-driven projects that improve efficiency and results." + }, + { + "span_id": "span-9404f81bce8d20931b66f045", + "text": "Yes: students get 50% off, previous book or course buyers are eligible for a discount, and groups of two or more get bundle pricing." + }, + { + "span_id": "span-b698c2919d8f70da2d89a9ae", + "text": "Contact us to claim." + }, + { + "span_id": "span-5cf2426a9ab03f1953edc553", + "text": "We offer course bundles and custom training tailored for companies, including live formats." + }, + { + "span_id": "span-ee13c693fe4e148b46bc0aa1", + "text": "See enterprise enablement or get in touch ." + }, + { + "span_id": "span-baaac8be6dd1825048f7ac39", + "text": "Lifetime access · $399" + } + ] + }, + { + "chunk_id": "tai-8e28f102910701825cf4ecdc", + "text": "Become the AI power user on your team\nBoost efficiency, enhance your team's capabilities, and spot the AI opportunities in your sector before your competitors do. Enroll now See the full pathway", + "heading": "Become the AI power user on your team", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-717beb62e72e0d678e12fb5d", + "text": "Boost efficiency, enhance your team's capabilities, and spot the AI opportunities in your sector before your competitors do." + }, + { + "span_id": "span-46ebcad1c0539c817ef16cab", + "text": "Enroll now See the full pathway" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "ad80d3bd9e702b03316286ed740aa0928ef9095b7338e8ea77826381ca294208", + "content_hash": "ad80d3bd9e702b03316286ed740aa0928ef9095b7338e8ea77826381ca294208", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/ai-for-work/" + ], + "evidence_hash": "7b9b164a00d3b8420a30f48d69f7f59c22a3f4982c03633fef6fea737a591190" + }, + { + "discovered_url": "https://towardsai.com/academy/building-llms-for-production/", + "url": "https://towardsai.com/academy/building-llms-for-production/", + "canonical_url": "https://towardsai.com/academy/building-llms-for-production/", + "host": "towardsai.com", + "path": "/academy/building-llms-for-production", + "kind": "book", + "offer_id": "building-llms-for-production", + "entity_id": "offer:building-llms-for-production", + "title": "Building LLMs for Production: The Ebook, Taught as a Course | Towards AI Academy", + "review_title": "Building LLMs for Production", + "reviewed_summary": "Towards AI book and companion learning page.", + "meta_description": "Our Building LLMs for Production ebook as an 84-lesson interactive course: prompting, RAG, fine-tuning, agents, and deployment, with code projects and Colab notebooks. Lifetime access for $29.99.", + "headings": [ + "Our Building LLMs for Production Ebook", + "What's inside this 470-page book (Updated October 2024)?", + "Industry leaders on the book", + "Every chapter, tutorial by tutorial", + "Building LLMs for Production", + "Who is it for?", + "What our readers say", + "The Only AI Engineering Toolkit You Need!", + "Everything the book covers", + "LLM Fundamentals, Architecture, & LLMs in Practice", + "Prompting & Frameworks", + "RAG & Fine-Tuning", + "Agents, Optimization & Deployment", + "Why should you read this book?", + "FAQ" + ], + "text": "Towards AI Academy · From our book\n\nOur Building LLMs for Production Ebook\nEnhancing LLM Abilities and Reliability with Prompting, Fine-Tuning, and RAG. Unlock Lifetime Access → $29.99 See what's inside ↓ With all future updates · 10,000+ copies sold · 84 lessons, taught as a course 01 Inside the book\n\nWhat's inside this 470-page book (Updated October 2024)?\nPlease note: this e-book is an interactive resource, not a downloadable PDF. 01 Hands-on Guide on LLMs, Prompting, Retrieval Augmented Generation (RAG) & Fine-tuning 02 Roadmap for Building Production-Ready Applications using LLMs 03 Fundamentals of LLM Theory 04 Simple-to-Advanced LLM Techniques & Frameworks 05 Code Projects with Real-World Applications 06 Colab Notebooks that you can run right away 07 Community access and our own AI Tutor 02 Trusted voices\n\nIndustry leaders on the book\n03 Chapter overview\n\nEvery chapter, tutorial by tutorial\n01 About the Book 2 items + 1 Table of Contents Free preview 2 About The Book Free preview 02 Introduction 3 items + 1 Introduction Free preview 2 Why Prompt Engineering, Fine-Tuning, and RAG? 3 Coding Environment and Packages 03 Chapter I: Introduction to LLMs 6 items + 1 A Brief History of Language Models 2 What are Large Language Models? 3 Building Blocks of LLMs 4 Tutorial: Translation with LLMs (GPT-3.5 API) 5 Tutorial: Control LLMs Output with Few-Shot Learning 6 Recap 04 Chapter II: LLM Architectures and Landscape 8 items + 1 Understanding Transformers 2 Transformer Model's Design Choices 3 Transformer Architecture Optimization Techniques 4 The Generative Pre-trained Transformer (GPT) Architecture 5 Introduction to Large Multimodal Models 6 Proprietary vs. Open Models vs. Open-Source Language Models 7 Applications and Use-Cases of LLMs 8 Recap 05 Chapter III: LLMs in Practice 4 items + 1 Understanding Hallucinations and Bias 2 Reducing Hallucinations by Controlling LLM Outputs 3 Evaluating LLM Performance 4 Recap 06 Chapter IV: Introduction to Prompting 4 items + 1 Prompting and Prompt Engineering 2 Prompting Techniques 3 Prompt Injection and Security 4 Recap 07 Chapter V: Retrieval-Augmented Generation 3 items + 1 Why RAG? 2 Building a Basic RAG Pipeline from Scratch 3 Recap 08 Chapter VI: Introduction to LangChain & LlamaIndex 7 items + 1 LLM Frameworks 2 LangChain Introduction 3 Tutorial 1: Building LLM-Powered Applications with LangChain 4 Tutorial 2: Building a News Articles Summarizer 5 LlamaIndex Introduction 6 LangChain vs. LlamaIndex vs. OpenAI Assistants 7 Recap 09 Chapter VII: Prompting with LangChain 7 items + 1 What are LangChain Prompt Templates 2 Few-Shot Prompts and Example Selectors 3 What are LangChain Chains 4 Tutorial 1: Managing Outputs with Output Parsers 5 Tutorial 2: Improving Our News Articles Summarizer 6 Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections 7 Recap 10 Chapter VIII: Indexes, Retrievers, and Data Preparation 10 items + 1 LangChain's Indexes and Retrievers 2 Data Ingestion 3 Text Splitters 4 Similarity Search and Vector Embeddings 5 Tutorial 1: A Customer Support Q&A Chatbot 6 Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain 7 Tutorial 3: A Voice Assistant for Your Knowledge Base 8 Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain 9 Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot 10 Recap 11 Chapter IX: Advanced RAG 5 items + 1 From Proof of Concept to Product: Challenges of RAG Systems 2 Advanced RAG Techniques with LlamaIndex 3 RAG - Metrics & Evaluation 4 LangChain LangSmith and LangChain Hub 5 Recap 12 Chapter X: Agents 9 items + 1 What are Agents: Large Models as Reasoning Engines 2 An Overview of AutoGPT and BabyAGI 3 The Agent Simulation Projects in LangChain 4 Tutorial 1: Building Agents for Analysis Report Creation 5 Tutorial 2: Query and Summarize a DB with LlamaIndex 6 Tutorial 3: Building Agents with OpenAI Assistants 7 Tutorial 4: LangChain OpenGPT 8 Tutorial 5: Multimodal Financial Document Analysis from PDFs 9 Recap 13 Chapter XI: Fine-Tuning 8 items + 1 Understanding Fine-Tuning 2 Low-Rank Adaptation (LoRA) 3 Tutorial 1: SFT with LoRA 4 Tutorial 2: Using SFT and LoRA for Financial Sentiment 5 Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data 6 Reinforcement Learning from Human Feedback 7 Tutorial 4: Improving LLMs with RLHF 8 Recap 14 Chapter XII: Deployment and Optimization 5 items + 1 Model Distillation and Teacher-Student Models 2 LLM Deployment Optimization: Quantization, Pruning, and Speculative Decoding 3 Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP) 4 Deploying Open-Source LLMs on Cloud Providers 5 Recap 15 Conclusion 3 items + 1 Conclusion 2 Further Reading and Courses 3 AI Tutor - Ask any question!\n\nBuilding LLMs for Production\n$29.99 5.0 (26 reviews) 84 lessons Unlock Lifetime Access With all future updates 04 For you\n\nWho is it for?\n01 AI Practitioners & Programmers Tinkerers 02 AI/ML Engineers & Computer Science Professionals 03 Students/Researchers & Job Seekers 05 Reader reviews\n\nWhat our readers say\n\nThe Only AI Engineering Toolkit You Need!\nTo build scalable and reliable products with LLMs Get Your Copy Now! 06 The full stack\n\nEverything the book covers\n\nLLM Fundamentals, Architecture, & LLMs in Practice\nBuilding blocks of LLMs language modeling, tokenization, embeddings, emergent abilities, scaling laws, context size… Transformer Architecture attention mechanism, design choices, encoder-only, decoder-only and encoder-decoder transformers, GPT Architecture, Masked Self-Attention, MinGPT Hallucinations & Biases mitigation strategies, controlling LLM outputs Decoding methods greedy search, sampling, beam search, top-k sampling, top-p sampling Objective functions and evaluation metrics perplexity metric and GLUE, SuperGLUE, BIG-Bench, HELM, FLASK Benchmarks…\n\nPrompting & Frameworks\nPrompting techniques zero-shot, in context, few-shot, role, chains, and chain-of-thought… Security Prompt Injection and Prompt Hacking LangChain prompt templates, output parsers, summarization chain, QA chains LlamaIndex vector stores, embeddings, data connectors, nodes, indexes\n\nRAG & Fine-Tuning\nData Ingestion PDFs, web pages, Google Drive, text splitters, LangChain Chains Embeddings Vector Stores with Activeloop's Deep Lake Querying in LlamaIndex query construction, expansion, transformation, splitting, customizing a retriever engine… Reranking Documents recursive, small-to-big RAG Metrics Mean Reciprocal Rank (MRR), Hit Rate, Mean Average Precision (MAP), and Normalized Discounted Cumulative Gain (NDCG)… Evaluation Tools evaluating with ragas, custom evaluation of RAG pipelines Fine-Tuning Optimization Techniques LoRA, QLoRA, supervised fine-tuning, SFT RLHF\n\nAgents, Optimization & Deployment\nAgents Using AutoGPT & BabyAGI with LangChain Agent Simulation Projects CAMEL, Generative Agents Building Agents LangGPT, OpenAI Assistants Optimization & Deployment challenges, quantization, pruning, distillation, cloud deployment, CPU and GPU optimization & deployment, creating APIs from open-source LLMs 07 Why this book\n\nWhy should you read this book?\nFuture-Proof Skills This book explores various methods to adapt \"foundational\" LLMs to specific tasks with enhanced accuracy, reliability, and scalability. It tackles the lack of reliability of \"out of the box\" LLMs by teaching the AI developer tech stack of the future; Prompting, Fine-Tuning, RAG, and Tools Use. Scalable Solutions The book aims to guide developers through creating LLM products ready for production, leveraging the potential of AI across various industries. It breaks down techniques that are scalable for enterprise-level workflows, helping both independent developers and small companies with limited resources create AI products that deliver value to paying customers. Practical Expertise for Everyone The book is for anyone who wants to build LLM products that can serve real use cases today. It comes with access to our webpage where we also share lots of additional up-to-date content, code, notebooks, and resources. However, the coding parts of the book is tailored for readers with an intermediate knowledge of Python. Start Reading! → $29.99 08 Questions\n\nFAQ\nWhat skills do I learn? + The book is packed with theories, concepts, projects, applications, and experience that you can confidently put on your CVs. You can add these skills straight into your resume: Large Language Models (LLMs) | LangChain | LlamaIndex | Vector databases | RAG | Prompting | Fine-tuning | Agents | Deployment & Deployment Optimizations | Creating chatbots | Chat with PDFs | Summarization | AI Assistants | RLHF What are the prerequisites to read the book? + The book is written for readers without prior knowledge of AI or NLP. It introduces topics from the ground up, aiming to help you feel comfortable using the power of AI in your next project or to elevate your current project to the next level. A basic understanding of Python helps comprehend the code and implementations, while advanced use cases of the coding techniques are explained in detail in the course. How do we make sure the book is not outdated? + We ensure the book remains relevant by focusing on the core principles of building production products with LLMs, which are foundational and transferable across generations of models. While the field is fast-evolving and new techniques will emerge, today's LLM developer stack will still be crucial for adapting future models to specific industries and data. Additionally, we provide access to an up-to-date webpage with extra content, code, notebooks, and resources, ensuring readers stay current with the latest advancements. Does it come with a physical copy? + No. This e-book version is hosted on the platform (not a pdf). You can purchase a soft or hard copy of the book on Amazon . If you have a physical copy, email Louis-François at louis@towardsai.net , and we'd be happy to give you a discount on the e-book! Do you have a referral or affiliate program? + If you refer three or more people, we’ll send you a physical copy of our book as a thank you! Additionally, we have an affiliate program for individuals with an audience. By joining, you can earn commissions for every successful referral made through your unique affiliate link. Please email Louis-François at louis@towardsai.net with proof of referral. Can I take this course within my company? + Yes! We offer both course bundles and custom training solutions tailored specifically for companies. For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net .", + "links": [ + { + "text": "Unlock Lifetime Access → $29.99", + "url": "https://academy.towardsai.net/cart/add_product/3147339?price_id=4025407" + }, + { + "text": "Table of Contents Free preview", + "url": "https://academy.towardsai.net/enroll/3147339?et=free_trial" + }, + { + "text": "About The Book Free preview", + "url": "https://academy.towardsai.net/enroll/3147339?et=free_trial" + }, + { + "text": "Introduction Free preview", + "url": "https://academy.towardsai.net/enroll/3147339?et=free_trial" + }, + { + "text": "Unlock Lifetime Access", + "url": "https://academy.towardsai.net/cart/add_product/3147339?price_id=4025407" + }, + { + "text": "Get Your Copy Now!", + "url": "https://academy.towardsai.net/cart/add_product/3147339?price_id=4025407" + }, + { + "text": "Start Reading! → $29.99", + "url": "https://academy.towardsai.net/cart/add_product/3147339?price_id=4025407" + }, + { + "text": "on Amazon", + "url": "https://amzn.to/4bqYU9b" + } + ], + "chunks": [ + { + "chunk_id": "tai-924e204289f030a2d0822899", + "text": "Towards AI Academy · From our book\n\nOur Building LLMs for Production Ebook\nEnhancing LLM Abilities and Reliability with Prompting, Fine-Tuning, and RAG. Unlock Lifetime Access → $29.99 See what's inside ↓ With all future updates · 10,000+ copies sold · 84 lessons, taught as a course 01 Inside the book\n\nWhat's inside this 470-page book (Updated October 2024)?\nPlease note: this e-book is an interactive resource, not a downloadable PDF. 01 Hands-on Guide on LLMs, Prompting, Retrieval Augmented Generation (RAG) & Fine-tuning 02 Roadmap for Building Production-Ready Applications using LLMs 03 Fundamentals of LLM Theory 04 Simple-to-Advanced LLM Techniques & Frameworks 05 Code Projects with Real-World Applications 06 Colab Notebooks that you can run right away 07 Community access and our own AI Tutor 02 Trusted voices\n\nIndustry leaders on the book\n03 Chapter overview", + "heading": "What's inside this 470-page book (Updated October 2024)?", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-d23b6c308d3f8b735cd1b192", + "text": "Towards AI Academy · From our book" + }, + { + "span_id": "span-b1bc1a1327c33ef73fed9dd6", + "text": "Enhancing LLM Abilities and Reliability with Prompting, Fine-Tuning, and RAG." + }, + { + "span_id": "span-e30977f4d06c62603101cb67", + "text": "Unlock Lifetime Access → $29.99 See what's inside ↓" + }, + { + "span_id": "span-064710faed5fd16086548165", + "text": "With all future updates · 10,000+ copies sold · 84 lessons, taught as a course" + }, + { + "span_id": "span-b0b11c6425beeaa3f449c920", + "text": "01 Inside the book" + }, + { + "span_id": "span-d8542e1306eacd052870a5f0", + "text": "What's inside this 470-page book (Updated October 2024)?" + }, + { + "span_id": "span-c67dbb96af5b64f47df63dab", + "text": "Please note: this e-book is an interactive resource, not a downloadable PDF." + }, + { + "span_id": "span-c97f6d65ee1d5e64d02d8480", + "text": "01 Hands-on Guide on LLMs, Prompting, Retrieval Augmented Generation (RAG) & Fine-tuning" + }, + { + "span_id": "span-33600261b5dbeacd6847291f", + "text": "02 Roadmap for Building Production-Ready Applications using LLMs" + }, + { + "span_id": "span-c92cc832a1cc4e730b559dde", + "text": "03 Fundamentals of LLM Theory" + }, + { + "span_id": "span-3dcab4c52e0cebe3df51fbd7", + "text": "04 Simple-to-Advanced LLM Techniques & Frameworks" + }, + { + "span_id": "span-dca4dd10efe04088ab0f1a22", + "text": "05 Code Projects with Real-World Applications" + }, + { + "span_id": "span-25fcabd7d8f151418adcda04", + "text": "06 Colab Notebooks that you can run right away" + }, + { + "span_id": "span-3a8bed1544b9e3873b2f73a9", + "text": "07 Community access and our own AI Tutor" + }, + { + "span_id": "span-b7bf7f5d8e2f2b9fc7c2ff85", + "text": "02 Trusted voices" + }, + { + "span_id": "span-5708e625e7e59ae4bdff9f57", + "text": "03 Chapter overview" + } + ] + }, + { + "chunk_id": "tai-6efd40514351db1a1d05dd5d", + "text": "Every chapter, tutorial by tutorial\n01 About the Book 2 items + 1 Table of Contents Free preview 2 About The Book Free preview 02 Introduction 3 items + 1 Introduction Free preview 2 Why Prompt Engineering, Fine-Tuning, and RAG? 3 Coding Environment and Packages 03 Chapter I: Introduction to LLMs 6 items + 1 A Brief History of Language Models 2 What are Large Language Models? 3 Building Blocks of LLMs 4 Tutorial: Translation with LLMs (GPT-3.5 API) 5 Tutorial: Control LLMs Output with Few-Shot Learning 6 Recap 04 Chapter II: LLM Architectures and Landscape 8 items + 1 Understanding Transformers 2 Transformer Model's Design Choices 3 Transformer Architecture Optimization Techniques 4 The Generative Pre-trained Transformer (GPT) Architecture 5 Introduction to Large Multimodal Models 6 Proprietary vs. Open Models vs. Open-Source Language Models 7 Applications and Use-Cases of LLMs 8 Recap 05 Chapter III: LLMs in Practice 4 items + 1 Understanding Hallucinations and Bias 2 Reducing Hallucinations by Controlling LLM Outputs 3 Evaluating LLM Performance 4 Recap 06 Chapter IV: Introduction to Prompting 4 items + 1 Prompting and Prompt Engineering 2 Prompting Techniques 3 Prompt Injection and Security 4 Recap 07 Chapter V: Retrieval-Augmented Generation 3 items + 1 Why RAG? 2 Building a Basic RAG Pipeline from Scratch 3 Recap 08 Chapter VI: Introduction to LangChain & LlamaIndex 7 items + 1 LLM Frameworks 2 LangChain Introduction 3 Tutorial 1: Building LLM-Powered Applications with LangChain 4 Tutorial 2: Building a News Articles Summarizer 5 LlamaIndex Introduction 6 LangChain vs. LlamaIndex vs. OpenAI Assistants 7 Recap 09 Chapter VII: Prompting with LangChain 7 items + 1 What are LangChain Prompt Templates 2 Few-Shot Prompts and Example Selectors 3 What are LangChain Chains", + "heading": "Every chapter, tutorial by tutorial", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-3b8c9243b6c1b6fe2753936e", + "text": "01 About the Book 2 items + 1 Table of Contents Free preview" + }, + { + "span_id": "span-3b18e5a605a4f50615540f47", + "text": "2 About The Book Free preview" + }, + { + "span_id": "span-e8070d36e261020aa2f37365", + "text": "02 Introduction 3 items + 1 Introduction Free preview" + }, + { + "span_id": "span-1fd99cc58a3af7440351570d", + "text": "2 Why Prompt Engineering, Fine-Tuning, and RAG?" + }, + { + "span_id": "span-70d776fe33b67f390be13b7b", + "text": "3 Coding Environment and Packages" + }, + { + "span_id": "span-3d5ecc2ed028f0b993e30971", + "text": "03 Chapter I: Introduction to LLMs 6 items + 1 A Brief History of Language Models" + }, + { + "span_id": "span-78f1178ec40a053844fd2458", + "text": "2 What are Large Language Models?" + }, + { + "span_id": "span-f112e6484895bd22afb83b2d", + "text": "3 Building Blocks of LLMs" + }, + { + "span_id": "span-96af8fde94363179e0341058", + "text": "4 Tutorial: Translation with LLMs (GPT-3.5 API)" + }, + { + "span_id": "span-e133180b5d0a9ce105508d36", + "text": "5 Tutorial: Control LLMs Output with Few-Shot Learning" + }, + { + "span_id": "span-c66647d8571fe442668fe6f8", + "text": "6 Recap" + }, + { + "span_id": "span-699c7d7a20224bfda7ccfd10", + "text": "04 Chapter II: LLM Architectures and Landscape 8 items + 1 Understanding Transformers" + }, + { + "span_id": "span-671072647ba080446ad5b706", + "text": "2 Transformer Model's Design Choices" + }, + { + "span_id": "span-8d07e4e3dc0a9da8871ca720", + "text": "3 Transformer Architecture Optimization Techniques" + }, + { + "span_id": "span-100d31cdad818493543486ff", + "text": "4 The Generative Pre-trained Transformer (GPT) Architecture" + }, + { + "span_id": "span-f628eb58c5669bc0a46768f1", + "text": "5 Introduction to Large Multimodal Models" + }, + { + "span_id": "span-bbcbed9629f2deea57327ca4", + "text": "6 Proprietary vs." + }, + { + "span_id": "span-35fa4b4a51e16df7ba632cbf", + "text": "Open Models vs." + }, + { + "span_id": "span-f1047a78a52b2ca940482484", + "text": "Open-Source Language Models" + }, + { + "span_id": "span-12314ae2fcc2618ed525dfed", + "text": "7 Applications and Use-Cases of LLMs" + }, + { + "span_id": "span-1231aa99f241a914f31a5d81", + "text": "8 Recap" + }, + { + "span_id": "span-662d78baee7140db1adb0532", + "text": "05 Chapter III: LLMs in Practice 4 items + 1 Understanding Hallucinations and Bias" + }, + { + "span_id": "span-e6bfcd09f130efe11cd845f7", + "text": "2 Reducing Hallucinations by Controlling LLM Outputs" + }, + { + "span_id": "span-2769fc34aebb6c7edabd0263", + "text": "3 Evaluating LLM Performance" + }, + { + "span_id": "span-61cc17f25fdbdc30c098416b", + "text": "4 Recap" + }, + { + "span_id": "span-68a1e2aefb46fc72ba9f09bb", + "text": "06 Chapter IV: Introduction to Prompting 4 items + 1 Prompting and Prompt Engineering" + }, + { + "span_id": "span-8a20af9de10c0bbb6b543346", + "text": "2 Prompting Techniques" + }, + { + "span_id": "span-e7422f775cb1adf60ffdca23", + "text": "3 Prompt Injection and Security" + }, + { + "span_id": "span-4146bf48112ccaea58c4744e", + "text": "07 Chapter V: Retrieval-Augmented Generation 3 items + 1 Why RAG?" + }, + { + "span_id": "span-18cc9f733606b396f22a4b79", + "text": "2 Building a Basic RAG Pipeline from Scratch" + }, + { + "span_id": "span-864bf2866077931740ed7d01", + "text": "3 Recap" + }, + { + "span_id": "span-c157fcf3662858fb51b78f3a", + "text": "08 Chapter VI: Introduction to LangChain & LlamaIndex 7 items + 1 LLM Frameworks" + }, + { + "span_id": "span-2a95d9afa4298d6253a19da5", + "text": "2 LangChain Introduction" + }, + { + "span_id": "span-54a699317ba972a7337db34a", + "text": "3 Tutorial 1: Building LLM-Powered Applications with LangChain" + }, + { + "span_id": "span-9ba5fa3705863c7ddc56f367", + "text": "4 Tutorial 2: Building a News Articles Summarizer" + }, + { + "span_id": "span-86cbc08e464ab1c104748386", + "text": "5 LlamaIndex Introduction" + }, + { + "span_id": "span-95150040159c67a6a674ed11", + "text": "6 LangChain vs." + }, + { + "span_id": "span-364661b1dee83781a784b917", + "text": "LlamaIndex vs." + }, + { + "span_id": "span-dcebcd923c1744c35f38da7f", + "text": "OpenAI Assistants" + }, + { + "span_id": "span-e2c0a7da467f24d507c78dde", + "text": "7 Recap" + }, + { + "span_id": "span-141551472e5b8f4359c361a3", + "text": "09 Chapter VII: Prompting with LangChain 7 items + 1 What are LangChain Prompt Templates" + }, + { + "span_id": "span-1f92699dc1793008a17f5456", + "text": "2 Few-Shot Prompts and Example Selectors" + }, + { + "span_id": "span-f4c19e3b7007c5aeab4c39f3", + "text": "3 What are LangChain Chains" + } + ] + }, + { + "chunk_id": "tai-1cedfe75f3e6b9666e268420", + "text": "Every chapter, tutorial by tutorial\n4 Tutorial 1: Managing Outputs with Output Parsers 5 Tutorial 2: Improving Our News Articles Summarizer 6 Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections 7 Recap 10 Chapter VIII: Indexes, Retrievers, and Data Preparation 10 items + 1 LangChain's Indexes and Retrievers 2 Data Ingestion 3 Text Splitters 4 Similarity Search and Vector Embeddings 5 Tutorial 1: A Customer Support Q&A Chatbot 6 Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain 7 Tutorial 3: A Voice Assistant for Your Knowledge Base 8 Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain 9 Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot 10 Recap 11 Chapter IX: Advanced RAG 5 items + 1 From Proof of Concept to Product: Challenges of RAG Systems 2 Advanced RAG Techniques with LlamaIndex 3 RAG - Metrics & Evaluation 4 LangChain LangSmith and LangChain Hub 5 Recap 12 Chapter X: Agents 9 items + 1 What are Agents: Large Models as Reasoning Engines 2 An Overview of AutoGPT and BabyAGI 3 The Agent Simulation Projects in LangChain 4 Tutorial 1: Building Agents for Analysis Report Creation 5 Tutorial 2: Query and Summarize a DB with LlamaIndex 6 Tutorial 3: Building Agents with OpenAI Assistants 7 Tutorial 4: LangChain OpenGPT 8 Tutorial 5: Multimodal Financial Document Analysis from PDFs 9 Recap 13 Chapter XI: Fine-Tuning 8 items + 1 Understanding Fine-Tuning 2 Low-Rank Adaptation (LoRA) 3 Tutorial 1: SFT with LoRA 4 Tutorial 2: Using SFT and LoRA for Financial Sentiment 5 Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data 6 Reinforcement Learning from Human Feedback 7 Tutorial 4: Improving LLMs with RLHF 8 Recap 14 Chapter XII: Deployment and Optimization 5 items + 1 Model", + "heading": "Every chapter, tutorial by tutorial", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-d882fd88635fbdd7bf450967", + "text": "8 Recap" + }, + { + "span_id": "span-e74d7b422699d9b3d2e6ef3f", + "text": "OpenAI Assistants" + }, + { + "span_id": "span-8cf948828adf9bd1167239e4", + "text": "7 Recap" + }, + { + "span_id": "span-0f7fc51867511133ad48f29b", + "text": "4 Tutorial 1: Managing Outputs with Output Parsers" + }, + { + "span_id": "span-48b996a478b1619907f2562f", + "text": "5 Tutorial 2: Improving Our News Articles Summarizer" + }, + { + "span_id": "span-2225519bf1618f087df7d06b", + "text": "6 Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections" + }, + { + "span_id": "span-788515cff572e7bf0a352bfc", + "text": "10 Chapter VIII: Indexes, Retrievers, and Data Preparation 10 items + 1 LangChain's Indexes and Retrievers" + }, + { + "span_id": "span-4af0fd1874d18bf913ce57a8", + "text": "2 Data Ingestion" + }, + { + "span_id": "span-000351856a424233f70f720c", + "text": "3 Text Splitters" + }, + { + "span_id": "span-d0ffc3598032f3d9663a0250", + "text": "4 Similarity Search and Vector Embeddings" + }, + { + "span_id": "span-0162c3b4a1738b7650014eb0", + "text": "5 Tutorial 1: A Customer Support Q&A Chatbot" + }, + { + "span_id": "span-5cef8d8b0631076ebdc0a865", + "text": "6 Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain" + }, + { + "span_id": "span-695cdd263d3182de56cfdb1d", + "text": "7 Tutorial 3: A Voice Assistant for Your Knowledge Base" + }, + { + "span_id": "span-aa9976a2915a8537aac56d9d", + "text": "8 Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain" + }, + { + "span_id": "span-78c44227f69e8eb381b6ffec", + "text": "9 Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot" + }, + { + "span_id": "span-df5cce9b93b5fdcc19d47c75", + "text": "10 Recap" + }, + { + "span_id": "span-f2d38e1af549c7a7a63a8fe2", + "text": "11 Chapter IX: Advanced RAG 5 items + 1 From Proof of Concept to Product: Challenges of RAG Systems" + }, + { + "span_id": "span-67dba67dacfce84b3da7ce9c", + "text": "2 Advanced RAG Techniques with LlamaIndex" + }, + { + "span_id": "span-172a0ef43870730bc6ed04e6", + "text": "3 RAG - Metrics & Evaluation" + }, + { + "span_id": "span-eac2934e17feb112f26591cd", + "text": "4 LangChain LangSmith and LangChain Hub" + }, + { + "span_id": "span-2b0b7b3ac6752a5af9cdf01b", + "text": "5 Recap" + }, + { + "span_id": "span-bda4b20ff127a9e8dba34d3b", + "text": "12 Chapter X: Agents 9 items + 1 What are Agents: Large Models as Reasoning Engines" + }, + { + "span_id": "span-39002c9d7184efbdd838c0cc", + "text": "2 An Overview of AutoGPT and BabyAGI" + }, + { + "span_id": "span-807e0afdd3479f47a108e847", + "text": "3 The Agent Simulation Projects in LangChain" + }, + { + "span_id": "span-3ee0731f97421be1f6a1c468", + "text": "4 Tutorial 1: Building Agents for Analysis Report Creation" + }, + { + "span_id": "span-bc3574dc8e91f8c63e152652", + "text": "5 Tutorial 2: Query and Summarize a DB with LlamaIndex" + }, + { + "span_id": "span-da2a67f19151c04a09e67752", + "text": "6 Tutorial 3: Building Agents with OpenAI Assistants" + }, + { + "span_id": "span-55663922081d5d2211da3e50", + "text": "7 Tutorial 4: LangChain OpenGPT" + }, + { + "span_id": "span-e49d3b37f80f270509a02f9a", + "text": "8 Tutorial 5: Multimodal Financial Document Analysis from PDFs" + }, + { + "span_id": "span-8af20465a6a12a260dfe7ee4", + "text": "9 Recap" + }, + { + "span_id": "span-cafa6d59790ee7fc596ac826", + "text": "13 Chapter XI: Fine-Tuning 8 items + 1 Understanding Fine-Tuning" + }, + { + "span_id": "span-56c2bc7e152d96acb1c59631", + "text": "2 Low-Rank Adaptation (LoRA)" + }, + { + "span_id": "span-24f165d9472c3033c9a1188c", + "text": "3 Tutorial 1: SFT with LoRA" + }, + { + "span_id": "span-a9e20c9ef69091fa7cd48372", + "text": "4 Tutorial 2: Using SFT and LoRA for Financial Sentiment" + }, + { + "span_id": "span-3309c093c8abecd9c8da1d69", + "text": "5 Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data" + }, + { + "span_id": "span-717486b3b616c4a4a0fe57a9", + "text": "6 Reinforcement Learning from Human Feedback" + }, + { + "span_id": "span-a9bafe1a16c8d02bc52eac40", + "text": "7 Tutorial 4: Improving LLMs with RLHF" + } + ] + }, + { + "chunk_id": "tai-c623cd6a46da2e4cc3eb6eaa", + "text": "Every chapter, tutorial by tutorial\nDistillation and Teacher-Student Models 2 LLM Deployment Optimization: Quantization, Pruning, and Speculative Decoding 3 Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP) 4 Deploying Open-Source LLMs on Cloud Providers 5 Recap 15 Conclusion 3 items + 1 Conclusion 2 Further Reading and Courses 3 AI Tutor - Ask any question!\n\nBuilding LLMs for Production\n$29.99 5.0 (26 reviews) 84 lessons Unlock Lifetime Access With all future updates 04 For you\n\nWho is it for?\n01 AI Practitioners & Programmers Tinkerers 02 AI/ML Engineers & Computer Science Professionals 03 Students/Researchers & Job Seekers 05 Reader reviews\n\nWhat our readers say\n\nThe Only AI Engineering Toolkit You Need!\nTo build scalable and reliable products with LLMs Get Your Copy Now! 06 The full stack\n\nEverything the book covers\n\nLLM Fundamentals, Architecture, & LLMs in Practice\nBuilding blocks of LLMs language modeling, tokenization, embeddings, emergent abilities, scaling laws, context size… Transformer Architecture attention mechanism, design choices, encoder-only, decoder-only and encoder-decoder transformers, GPT Architecture, Masked Self-Attention, MinGPT Hallucinations & Biases mitigation strategies, controlling LLM outputs Decoding methods greedy search, sampling, beam search, top-k sampling, top-p sampling Objective functions and evaluation metrics perplexity metric and GLUE, SuperGLUE, BIG-Bench, HELM, FLASK Benchmarks…\n\nPrompting & Frameworks\nPrompting techniques zero-shot, in context, few-shot, role, chains, and chain-of-thought… Security Prompt Injection and Prompt Hacking LangChain prompt templates, output parsers, summarization chain, QA chains LlamaIndex vector stores, embeddings, data connectors, nodes, indexes", + "heading": "LLM Fundamentals, Architecture, & LLMs in Practice", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-7b5af0e5702214800c34de47", + "text": "5 Recap" + }, + { + "span_id": "span-da9adb076ed92c6d2cf98e1c", + "text": "2 LLM Deployment Optimization: Quantization, Pruning, and Speculative Decoding" + }, + { + "span_id": "span-97929e3977a191388cc66bd0", + "text": "3 Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP)" + }, + { + "span_id": "span-d5b9b3ce9b1ca34bc898e6bc", + "text": "4 Deploying Open-Source LLMs on Cloud Providers" + }, + { + "span_id": "span-720e64cbb05899b61a89559c", + "text": "15 Conclusion 3 items + 1 Conclusion" + }, + { + "span_id": "span-a1ece2d81ec4e635cac2fb32", + "text": "2 Further Reading and Courses" + }, + { + "span_id": "span-ef7e9074806c693103109e3d", + "text": "3 AI Tutor - Ask any question!" + }, + { + "span_id": "span-fccf0437a9d5cb2f0bca8f4a", + "text": "$29.99" + }, + { + "span_id": "span-fb57da6f2956f7f41cda9e78", + "text": "5.0 (26 reviews) 84 lessons" + }, + { + "span_id": "span-1af1459d341f5368fe260e04", + "text": "Unlock Lifetime Access With all future updates" + }, + { + "span_id": "span-4844842950839d30451717d2", + "text": "04 For you" + }, + { + "span_id": "span-456ec25c7a8103586a023d80", + "text": "01 AI Practitioners & Programmers Tinkerers" + }, + { + "span_id": "span-bbbc87cbe030630f4f43fc98", + "text": "02 AI/ML Engineers & Computer Science Professionals" + }, + { + "span_id": "span-c41763034bb171d9b9f4138f", + "text": "03 Students/Researchers & Job Seekers" + }, + { + "span_id": "span-da5e7ec8fdeee0b132af49b5", + "text": "05 Reader reviews" + }, + { + "span_id": "span-32410cd5af3a0a364e0477ad", + "text": "What our readers say" + }, + { + "span_id": "span-99dc941e2d8603480ea328d1", + "text": "To build scalable and reliable products with LLMs" + }, + { + "span_id": "span-b4d844e3d5ed0f334f96cca0", + "text": "Get Your Copy Now!" + }, + { + "span_id": "span-c9a6dce96a2cabf54908ede4", + "text": "06 The full stack" + }, + { + "span_id": "span-7e6a9f0f370ec4521e96336c", + "text": "Everything the book covers" + }, + { + "span_id": "span-7df784d57e96e3f6efedaf2c", + "text": "Building blocks of LLMs language modeling, tokenization, embeddings, emergent abilities, scaling laws, context size…" + }, + { + "span_id": "span-81dfd5c8bf930a80a3e9b708", + "text": "Transformer Architecture attention mechanism, design choices, encoder-only, decoder-only and encoder-decoder transformers, GPT Architecture, Masked Self-Attention, MinGPT" + }, + { + "span_id": "span-e7e4c5020e2b14b4f363cd21", + "text": "Hallucinations & Biases mitigation strategies, controlling LLM outputs" + }, + { + "span_id": "span-521d0abe85a66545b877fe20", + "text": "Decoding methods greedy search, sampling, beam search, top-k sampling, top-p sampling" + }, + { + "span_id": "span-d490efc542b9a034eb5be2d1", + "text": "Objective functions and evaluation metrics perplexity metric and GLUE, SuperGLUE, BIG-Bench, HELM, FLASK Benchmarks…" + }, + { + "span_id": "span-15146b30e35994bcaae14b4b", + "text": "Prompting techniques zero-shot, in context, few-shot, role, chains, and chain-of-thought…" + }, + { + "span_id": "span-5e20976e9d20ddfebfdf64a7", + "text": "Security Prompt Injection and Prompt Hacking" + }, + { + "span_id": "span-9ca171c81f5a9d132ce59787", + "text": "LangChain prompt templates, output parsers, summarization chain, QA chains" + }, + { + "span_id": "span-b655068208dd240537c8326e", + "text": "LlamaIndex vector stores, embeddings, data connectors, nodes, indexes" + } + ] + }, + { + "chunk_id": "tai-6e196f137332f08daf852c2a", + "text": "RAG & Fine-Tuning\nData Ingestion PDFs, web pages, Google Drive, text splitters, LangChain Chains Embeddings Vector Stores with Activeloop's Deep Lake Querying in LlamaIndex query construction, expansion, transformation, splitting, customizing a retriever engine… Reranking Documents recursive, small-to-big RAG Metrics Mean Reciprocal Rank (MRR), Hit Rate, Mean Average Precision (MAP), and Normalized Discounted Cumulative Gain (NDCG)… Evaluation Tools evaluating with ragas, custom evaluation of RAG pipelines Fine-Tuning Optimization Techniques LoRA, QLoRA, supervised fine-tuning, SFT RLHF\n\nAgents, Optimization & Deployment\nAgents Using AutoGPT & BabyAGI with LangChain Agent Simulation Projects CAMEL, Generative Agents Building Agents LangGPT, OpenAI Assistants Optimization & Deployment challenges, quantization, pruning, distillation, cloud deployment, CPU and GPU optimization & deployment, creating APIs from open-source LLMs 07 Why this book", + "heading": "RAG & Fine-Tuning", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-2290891f9a273b18bf0c12e5", + "text": "Data Ingestion PDFs, web pages, Google Drive, text splitters, LangChain Chains" + }, + { + "span_id": "span-2e510adc01800fa3510a9ea8", + "text": "Embeddings Vector Stores with Activeloop's Deep Lake" + }, + { + "span_id": "span-5f279c67077e4ce17669f992", + "text": "Querying in LlamaIndex query construction, expansion, transformation, splitting, customizing a retriever engine…" + }, + { + "span_id": "span-89ba7358eb66e755d2edd754", + "text": "Reranking Documents recursive, small-to-big" + }, + { + "span_id": "span-1bb298ca31df8d94acbf4584", + "text": "RAG Metrics Mean Reciprocal Rank (MRR), Hit Rate, Mean Average Precision (MAP), and Normalized Discounted Cumulative Gain (NDCG)…" + }, + { + "span_id": "span-75b45c2f42f825c9f1ad8716", + "text": "Evaluation Tools evaluating with ragas, custom evaluation of RAG pipelines" + }, + { + "span_id": "span-b31523106e767f71bf970173", + "text": "Fine-Tuning Optimization Techniques LoRA, QLoRA, supervised fine-tuning, SFT RLHF" + }, + { + "span_id": "span-3c1dcaae975105711cc17d46", + "text": "Agents Using AutoGPT & BabyAGI with LangChain" + }, + { + "span_id": "span-c779cb8443ec7693282fd576", + "text": "Agent Simulation Projects CAMEL, Generative Agents" + }, + { + "span_id": "span-589ce0caec71d48a8ddea8bf", + "text": "Building Agents LangGPT, OpenAI Assistants" + }, + { + "span_id": "span-2147f822d54a836cf7bc031f", + "text": "Optimization & Deployment challenges, quantization, pruning, distillation, cloud deployment, CPU and GPU optimization & deployment, creating APIs from open-source LLMs" + }, + { + "span_id": "span-dc81e14f7e154a7639735470", + "text": "07 Why this book" + } + ] + }, + { + "chunk_id": "tai-8e75dc26d105e111586b68b5", + "text": "Why should you read this book?\nFuture-Proof Skills This book explores various methods to adapt \"foundational\" LLMs to specific tasks with enhanced accuracy, reliability, and scalability. It tackles the lack of reliability of \"out of the box\" LLMs by teaching the AI developer tech stack of the future; Prompting, Fine-Tuning, RAG, and Tools Use. Scalable Solutions The book aims to guide developers through creating LLM products ready for production, leveraging the potential of AI across various industries. It breaks down techniques that are scalable for enterprise-level workflows, helping both independent developers and small companies with limited resources create AI products that deliver value to paying customers. Practical Expertise for Everyone The book is for anyone who wants to build LLM products that can serve real use cases today. It comes with access to our webpage where we also share lots of additional up-to-date content, code, notebooks, and resources. However, the coding parts of the book is tailored for readers with an intermediate knowledge of Python. Start Reading! → $29.99 08 Questions", + "heading": "Why should you read this book?", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-38ed9acf5cc380771c6eb630", + "text": "Future-Proof Skills This book explores various methods to adapt \"foundational\" LLMs to specific tasks with enhanced accuracy, reliability, and scalability." + }, + { + "span_id": "span-72e074d8e150682ce775f9bb", + "text": "It tackles the lack of reliability of \"out of the box\" LLMs by teaching the AI developer tech stack of the future; Prompting, Fine-Tuning, RAG, and Tools Use." + }, + { + "span_id": "span-3a7bc5f9dae66c061b6b2e1d", + "text": "Scalable Solutions The book aims to guide developers through creating LLM products ready for production, leveraging the potential of AI across various industries." + }, + { + "span_id": "span-daa98e4e7551034d2536394d", + "text": "It breaks down techniques that are scalable for enterprise-level workflows, helping both independent developers and small companies with limited resources create AI products that deliver value to paying customers." + }, + { + "span_id": "span-0399da596aae1364178f3e6d", + "text": "Practical Expertise for Everyone The book is for anyone who wants to build LLM products that can serve real use cases today." + }, + { + "span_id": "span-764957da66fd79505871811c", + "text": "It comes with access to our webpage where we also share lots of additional up-to-date content, code, notebooks, and resources." + }, + { + "span_id": "span-413e5d4aa77643bb03adc4cb", + "text": "However, the coding parts of the book is tailored for readers with an intermediate knowledge of Python." + }, + { + "span_id": "span-c6edd7717d3d3f3ca4952cc3", + "text": "Start Reading! → $29.99" + }, + { + "span_id": "span-52d99de76e547cc276947a25", + "text": "08 Questions" + } + ] + }, + { + "chunk_id": "tai-67ecaf5a28dabcdf0d2be511", + "text": "FAQ\nWhat skills do I learn? + The book is packed with theories, concepts, projects, applications, and experience that you can confidently put on your CVs. You can add these skills straight into your resume: Large Language Models (LLMs) | LangChain | LlamaIndex | Vector databases | RAG | Prompting | Fine-tuning | Agents | Deployment & Deployment Optimizations | Creating chatbots | Chat with PDFs | Summarization | AI Assistants | RLHF What are the prerequisites to read the book? + The book is written for readers without prior knowledge of AI or NLP. It introduces topics from the ground up, aiming to help you feel comfortable using the power of AI in your next project or to elevate your current project to the next level. A basic understanding of Python helps comprehend the code and implementations, while advanced use cases of the coding techniques are explained in detail in the course. How do we make sure the book is not outdated? + We ensure the book remains relevant by focusing on the core principles of building production products with LLMs, which are foundational and transferable across generations of models. While the field is fast-evolving and new techniques will emerge, today's LLM developer stack will still be crucial for adapting future models to specific industries and data. Additionally, we provide access to an up-to-date webpage with extra content, code, notebooks, and resources, ensuring readers stay current with the latest advancements. Does it come with a physical copy? + No. This e-book version is hosted on the platform (not a pdf). You can purchase a soft or hard copy of the book on Amazon . If you have a physical copy, email Louis-François at louis@towardsai.net , and we'd be happy to give you a discount on the e-book! Do you have a referral or affiliate", + "heading": "FAQ", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-cb92cf0b4fc8684c1f656037", + "text": "This e-book version is hosted on the platform (not a pdf)." + }, + { + "span_id": "span-01d89dd3d8abd6c33a8dd059", + "text": "You can purchase a soft or hard copy of the book on Amazon ." + }, + { + "span_id": "span-0e4546af14b930b533bf3023", + "text": "If you have a physical copy, email Louis-François at louis@towardsai.net , and we'd be happy to give you a discount on the e-book!" + } + ] + }, + { + "chunk_id": "tai-da410cd27bfd4926906f72a1", + "text": "FAQ\nprogram? + If you refer three or more people, we’ll send you a physical copy of our book as a thank you! Additionally, we have an affiliate program for individuals with an audience. By joining, you can earn commissions for every successful referral made through your unique affiliate link. Please email Louis-François at louis@towardsai.net with proof of referral. Can I take this course within my company? + Yes! We offer both course bundles and custom training solutions tailored specifically for companies. For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net .", + "heading": "FAQ", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-8ee9955fd96e38712c47ae0c", + "text": "We offer both course bundles and custom training solutions tailored specifically for companies." + }, + { + "span_id": "span-e1540888495821c66ae36528", + "text": "For more information on company packages or to discuss a customized training plan, reach out to Louis at louis@towardsai.net ." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "01e022df0186347b8bc459c4ded806f75b56f27f6b528098420fd6732e8b996e", + "content_hash": "01e022df0186347b8bc459c4ded806f75b56f27f6b528098420fd6732e8b996e", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/building-llms-for-production/" + ], + "evidence_hash": "e841423e5792401a7480f60e29acff98fe5e0a748328a5dd30d29809d5438684" + }, + { + "discovered_url": "https://towardsai.com/academy/bundles/", + "url": "https://towardsai.com/academy/bundles/", + "canonical_url": "https://towardsai.com/academy/bundles/", + "host": "towardsai.com", + "path": "/academy/bundles", + "kind": "collection", + "offer_id": null, + "entity_id": "page:towardsai.com/academy/bundles", + "title": "Our Courses and Bundles · The Full Towards AI Academy Catalog", + "review_title": "Academy bundles", + "reviewed_summary": "Central collection of current Academy bundles.", + "meta_description": "Every Towards AI Academy product on one page, with a two-question course finder: six self-paced AI courses, the engineer-led mentorship, and three discounted bundles, from beginner Python to production agent engineering.", + "headings": [ + "Our Courses and Bundles", + "Two questions, one recommendation", + "Six courses, from first Python script to production agents", + "Full Stack AI Engineering", + "Agent Engineering", + "10-Hour LLM Fundamentals", + "Beginner Python for AI Engineering", + "Master AI for Work", + "Building LLMs for Production", + "A senior AI team behind you", + "Career support and production mentorship from the Towards AI engineering team", + "Three bundles, one checkout", + "From Non-Coder to AI Engineer", + "From Developer to Advanced AI Engineer", + "Get It All: From Novice to Expert", + "Not sure where to start?" + ], + "text": "Towards AI Academy · Full catalog\n\nOur Courses and Bundles\nWhatever you're building toward, the route is here: six hands-on courses, senior AI engineers on call in the mentorship, and three bundles that cut the price of the full path. Find your path ↓ Browse everything ↓ Self-paced with lifetime access · Taught by practicing AI engineers 01 Find your path\n\nTwo questions, one recommendation\nEvery path below is self-paced with lifetime access. Tell us where you're starting and where you're headed, and we'll point at the option we'd pick in your shoes. 01 · Where are you starting from? I don't code New to coding Some Python Experienced developer 02 · What do you want from AI? Get more done at work LLM foundations, fast Become an AI engineer Ship production agents Our recommendation Lifetime access · certification · 30-day money-back guarantee Want senior engineers on call while you learn? The Mentorship adds live Q&A, written answers, and expert reviews for $99/month. 02 The courses\n\nSix courses, from first Python script to production agents\nProject-based, kept current by the engineers who teach them, and rated by the students who took them. Course · Flagship certification · step 1 of the engineer route\n\nFull Stack AI Engineering\n5.0 · 43 reviews Our flagship program. Build a working full-stack LLM product and master the skills companies need, through advanced RAG, fine-tuning, agents, and deployment. $349 View course → Course · Advanced · the follow-on to Full Stack\n\nAgent Engineering\n5.0 · 26 reviews Design, build, evaluate, and deploy production-grade AI agents and multi-agent systems. Made for developers tasked with shipping autonomous features. $499 View course → Course · Video crash course\n\n10-Hour LLM Fundamentals\n5.0 · 7 reviews The fastest route to LLM literacy: prompting, RAG, evaluation, agents, and optimization in ten hours of video. Know what to use, when, and why. $199 View course → Course · For beginners\n\nBeginner Python for AI Engineering\n5.0 · 4 reviews A Python course built for the LLM era. Learn to code by building toward your own AI applications, no programming background required. $149 View course → Course · No-code, for professionals\n\nMaster AI for Work\n5.0 · 2 reviews Real productivity gains from AI without writing code, with role-specific modules for sales, marketing, finance, HR, and leadership. $399 View course → Course · From our book\n\nBuilding LLMs for Production\n5.0 · 26 reviews Our book (10,000+ copies sold), taught as an 84-lesson course: scalable LLM techniques for developers and small teams building sellable AI products. $29.99 View course → 03 The mentorship\n\nA senior AI team behind you\nTowards AI Mentorship · membership\n\nCareer support and production mentorship from the Towards AI engineering team\nLive Q&A twice a week, async answers from senior engineers any day, expert reviews of your resume and projects, and the 10-Hour LLM Fundamentals course included. For the questions a course can't answer. $99 /month Explore the mentorship → 04 The bundles\n\nThree bundles, one checkout\nCurated course sequences at a lower combined price, matched to where you're starting from. Bundle · 4 learning products\n\nFrom Non-Coder to AI Engineer\nStart from zero. Beginner Python, the 10-Hour LLM Fundamentals video course, Full Stack AI Engineering, and the Building LLMs for Production e-book. $727 $599 Save $128 View bundle → Bundle · 4 learning products\n\nFrom Developer to Advanced AI Engineer\nAlready code? The 10-Hour LLM Fundamentals crash course, Full Stack AI Engineering, Agent Engineering, and the e-book: the road to senior AI capability. $1,078 $948 Save $130 View bundle → Bundle · 6 learning products\n\nGet It All: From Novice to Expert\nThe complete curriculum for anyone: every course above plus the e-book, from your first line of Python to certified agent engineering. $1,575 $1,349 Save $226 View bundle →\n\nNot sure where to start?\nAnswer the two questions above for our recommendation, and our team reads every message. Find your path Talk to us Prices in USD · Students get 50% off all courses", + "links": [ + { + "text": "Mentorship", + "url": "https://towardsai.com/academy/mentorship/" + }, + { + "text": "Course · Flagship certification · step 1 of the engineer route 5.0 · 43 reviews Our flagship program. Build a working full-stack LLM product and master the skills companies need, through advanced RAG, fine-tuning, agents, and deployment. $349 View course →", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Course · Advanced · the follow-on to Full Stack 5.0 · 26 reviews Design, build, evaluate, and deploy production-grade AI agents and multi-agent systems. Made for developers tasked with shipping autonomous features. $499 View course →", + "url": "https://towardsai.com/academy/agent-engineering/" + }, + { + "text": "Course · Video crash course 5.0 · 7 reviews The fastest route to LLM literacy: prompting, RAG, evaluation, agents, and optimization in ten hours of video. Know what to use, when, and why. $199 View course →", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "Course · For beginners 5.0 · 4 reviews A Python course built for the LLM era. Learn to code by building toward your own AI applications, no programming background required. $149 View course →", + "url": "https://towardsai.com/academy/python-for-ai-engineering/" + }, + { + "text": "Course · No-code, for professionals 5.0 · 2 reviews Real productivity gains from AI without writing code, with role-specific modules for sales, marketing, finance, HR, and leadership. $399 View course →", + "url": "https://towardsai.com/academy/ai-for-work/" + }, + { + "text": "Course · From our book 5.0 · 26 reviews Our book (10,000+ copies sold), taught as an 84-lesson course: scalable LLM techniques for developers and small teams building sellable AI products. $29.99 View course →", + "url": "https://towardsai.com/academy/building-llms-for-production/" + }, + { + "text": "Explore the mentorship →", + "url": "https://towardsai.com/academy/mentorship/" + }, + { + "text": "Bundle · 4 learning products Start from zero. Beginner Python, the 10-Hour LLM Fundamentals video course, Full Stack AI Engineering, and the Building LLMs for Production e-book. $727 $599 Save $128 View bundle →", + "url": "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/" + }, + { + "text": "Bundle · 4 learning products Already code? The 10-Hour LLM Fundamentals crash course, Full Stack AI Engineering, Agent Engineering, and the e-book: the road to senior AI capability. $1,078 $948 Save $130 View bundle →", + "url": "https://towardsai.com/academy/bundles/10-hour-crash-course-into-llm-developer-expert/" + }, + { + "text": "Bundle · 6 learning products The complete curriculum for anyone: every course above plus the e-book, from your first line of Python to certified agent engineering. $1,575 $1,349 Save $226 View bundle →", + "url": "https://towardsai.com/academy/bundles/get-it-all/" + }, + { + "text": "Talk to us", + "url": "https://towardsai.com/academy/contact/" + } + ], + "chunks": [ + { + "chunk_id": "tai-bc19fda0474ce8e61fb527f0", + "text": "Towards AI Academy · Full catalog\n\nOur Courses and Bundles\nWhatever you're building toward, the route is here: six hands-on courses, senior AI engineers on call in the mentorship, and three bundles that cut the price of the full path. Find your path ↓ Browse everything ↓ Self-paced with lifetime access · Taught by practicing AI engineers 01 Find your path\n\nTwo questions, one recommendation\nEvery path below is self-paced with lifetime access. Tell us where you're starting and where you're headed, and we'll point at the option we'd pick in your shoes. 01 · Where are you starting from? I don't code New to coding Some Python Experienced developer 02 · What do you want from AI? Get more done at work LLM foundations, fast Become an AI engineer Ship production agents Our recommendation Lifetime access · certification · 30-day money-back guarantee Want senior engineers on call while you learn? The Mentorship adds live Q&A, written answers, and expert reviews for $99/month. 02 The courses\n\nSix courses, from first Python script to production agents\nProject-based, kept current by the engineers who teach them, and rated by the students who took them. Course · Flagship certification · step 1 of the engineer route\n\nFull Stack AI Engineering\n5.0 · 43 reviews Our flagship program. Build a working full-stack LLM product and master the skills companies need, through advanced RAG, fine-tuning, agents, and deployment. $349 View course → Course · Advanced · the follow-on to Full Stack\n\nAgent Engineering\n5.0 · 26 reviews Design, build, evaluate, and deploy production-grade AI agents and multi-agent systems. Made for developers tasked with shipping autonomous features. $499 View course → Course · Video crash course", + "heading": "Two questions, one recommendation", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-e0ba41454d63fc1fce8f045e", + "text": "Towards AI Academy · Full catalog" + }, + { + "span_id": "span-afe83fa0a293a3ac6a22e97f", + "text": "Whatever you're building toward, the route is here: six hands-on courses, senior AI engineers on call in the mentorship, and three bundles that cut the price of the full path." + }, + { + "span_id": "span-a514f8b0fb3a210d1bd48b4b", + "text": "Find your path ↓ Browse everything ↓" + }, + { + "span_id": "span-e1e62a6f3805811090e5cc7b", + "text": "Self-paced with lifetime access · Taught by practicing AI engineers" + }, + { + "span_id": "span-3defb6de8f6e894ba0f14e6b", + "text": "01 Find your path" + }, + { + "span_id": "span-9d2e384f380c83203e982252", + "text": "Every path below is self-paced with lifetime access." + }, + { + "span_id": "span-e94b8e3b0f91428a7f9232bf", + "text": "Tell us where you're starting and where you're headed, and we'll point at the option we'd pick in your shoes." + }, + { + "span_id": "span-15e4e7317020d24cb3a71843", + "text": "01 · Where are you starting from?" + }, + { + "span_id": "span-944fbf828d14bf66e1933dc3", + "text": "I don't code New to coding Some Python Experienced developer" + }, + { + "span_id": "span-a04c3badc0ffff92d3f12f3b", + "text": "02 · What do you want from AI?" + }, + { + "span_id": "span-5026e971f49077cb54885036", + "text": "Get more done at work LLM foundations, fast Become an AI engineer Ship production agents" + }, + { + "span_id": "span-1b2c7d84550ea2f327f721b2", + "text": "Our recommendation" + }, + { + "span_id": "span-5b9463805500835166f4cbbb", + "text": "Lifetime access · certification · 30-day money-back guarantee" + }, + { + "span_id": "span-c0858e5d10b5d41a3916d286", + "text": "Want senior engineers on call while you learn?" + }, + { + "span_id": "span-0c304598fe1b029fd4c04862", + "text": "The Mentorship adds live Q&A, written answers, and expert reviews for $99/month." + }, + { + "span_id": "span-b6fe0659707cb72cefa07d7e", + "text": "02 The courses" + }, + { + "span_id": "span-44116bc4dd7a870294c3d544", + "text": "Project-based, kept current by the engineers who teach them, and rated by the students who took them." + }, + { + "span_id": "span-82a285c51adb60309374aa36", + "text": "Course · Flagship certification · step 1 of the engineer route" + }, + { + "span_id": "span-968f44a4c8998867eb00e2c2", + "text": "5.0 · 43 reviews" + }, + { + "span_id": "span-0365f95c97075014d4c15c49", + "text": "Our flagship program." + }, + { + "span_id": "span-20b989a0ab62b1b678fe8a36", + "text": "Build a working full-stack LLM product and master the skills companies need, through advanced RAG, fine-tuning, agents, and deployment." + }, + { + "span_id": "span-aeda15ba4be7f28a075c0432", + "text": "$349 View course →" + }, + { + "span_id": "span-7cdd89b9900fa4846b91fd6f", + "text": "Course · Advanced · the follow-on to Full Stack" + }, + { + "span_id": "span-24ebd567cce559fc7753a81b", + "text": "5.0 · 26 reviews" + }, + { + "span_id": "span-d5d669f9a2dc655f8182b236", + "text": "Design, build, evaluate, and deploy production-grade AI agents and multi-agent systems." + }, + { + "span_id": "span-11c6df9ac329d9f12fe1c3fb", + "text": "Made for developers tasked with shipping autonomous features." + }, + { + "span_id": "span-67aa601cc7c12ebc6a22fda3", + "text": "$499 View course →" + }, + { + "span_id": "span-6f9e06b61ff1372f0c63e1eb", + "text": "Course · Video crash course" + } + ] + }, + { + "chunk_id": "tai-688b351c0c7e928a484530ba", + "text": "10-Hour LLM Fundamentals\n5.0 · 7 reviews The fastest route to LLM literacy: prompting, RAG, evaluation, agents, and optimization in ten hours of video. Know what to use, when, and why. $199 View course → Course · For beginners\n\nBeginner Python for AI Engineering\n5.0 · 4 reviews A Python course built for the LLM era. Learn to code by building toward your own AI applications, no programming background required. $149 View course → Course · No-code, for professionals\n\nMaster AI for Work\n5.0 · 2 reviews Real productivity gains from AI without writing code, with role-specific modules for sales, marketing, finance, HR, and leadership. $399 View course → Course · From our book\n\nBuilding LLMs for Production\n5.0 · 26 reviews Our book (10,000+ copies sold), taught as an 84-lesson course: scalable LLM techniques for developers and small teams building sellable AI products. $29.99 View course → 03 The mentorship\n\nA senior AI team behind you\nTowards AI Mentorship · membership\n\nCareer support and production mentorship from the Towards AI engineering team\nLive Q&A twice a week, async answers from senior engineers any day, expert reviews of your resume and projects, and the 10-Hour LLM Fundamentals course included. For the questions a course can't answer. $99 /month Explore the mentorship → 04 The bundles\n\nThree bundles, one checkout\nCurated course sequences at a lower combined price, matched to where you're starting from. Bundle · 4 learning products\n\nFrom Non-Coder to AI Engineer\nStart from zero. Beginner Python, the 10-Hour LLM Fundamentals video course, Full Stack AI Engineering, and the Building LLMs for Production e-book. $727 $599 Save $128 View bundle → Bundle · 4 learning products", + "heading": "Career support and production mentorship from the Towards AI engineering team", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-9d4671b49d53527bb4750c26", + "text": "10-Hour LLM Fundamentals" + }, + { + "span_id": "span-2ac45eea2b9a2960b7e12cd5", + "text": "5.0 · 7 reviews" + }, + { + "span_id": "span-a4925d99356a8a8f017cf26f", + "text": "The fastest route to LLM literacy: prompting, RAG, evaluation, agents, and optimization in ten hours of video." + }, + { + "span_id": "span-6a4b9c592ecafcb6921d171b", + "text": "Know what to use, when, and why." + }, + { + "span_id": "span-5d2b53fb5abdc9dffe56bda6", + "text": "$199 View course →" + }, + { + "span_id": "span-85701861f370e0819b639aab", + "text": "Course · For beginners" + }, + { + "span_id": "span-62d7fb04736d3950d5806c1e", + "text": "5.0 · 4 reviews" + }, + { + "span_id": "span-a3030c344cd32eb04e82448b", + "text": "A Python course built for the LLM era." + }, + { + "span_id": "span-d70527c390865d57abd6d201", + "text": "Learn to code by building toward your own AI applications, no programming background required." + }, + { + "span_id": "span-dd2dbdd0971b60cbe28fa029", + "text": "$149 View course →" + }, + { + "span_id": "span-b1c946a528a8a8d7b0afada8", + "text": "Course · No-code, for professionals" + }, + { + "span_id": "span-0e6f1bb4f75b37a68893c64b", + "text": "5.0 · 2 reviews" + }, + { + "span_id": "span-5c584d7f31900b7c49f46310", + "text": "Real productivity gains from AI without writing code, with role-specific modules for sales, marketing, finance, HR, and leadership." + }, + { + "span_id": "span-1cf6e105da5d8e29833c1bdf", + "text": "$399 View course →" + }, + { + "span_id": "span-379ef2619a5f98b137cdc243", + "text": "Course · From our book" + }, + { + "span_id": "span-747c64e17c49bc3dc35ba3c8", + "text": "5.0 · 26 reviews" + }, + { + "span_id": "span-f0961905a15063056e8d66ab", + "text": "Our book (10,000+ copies sold), taught as an 84-lesson course: scalable LLM techniques for developers and small teams building sellable AI products." + }, + { + "span_id": "span-3268b1c7147ef8c2689613a1", + "text": "$29.99 View course →" + }, + { + "span_id": "span-61e18bedb81deb0e8a72d9ec", + "text": "03 The mentorship" + }, + { + "span_id": "span-0d0c32a46592c2bf7fd60f29", + "text": "Towards AI Mentorship · membership" + }, + { + "span_id": "span-a581e8e06eb3c0ba403c5298", + "text": "Live Q&A twice a week, async answers from senior engineers any day, expert reviews of your resume and projects, and the 10-Hour LLM Fundamentals course included." + }, + { + "span_id": "span-7028e8c2c808b1cd982cfeea", + "text": "For the questions a course can't answer." + }, + { + "span_id": "span-c14c1b5d1143815be0574509", + "text": "$99 /month Explore the mentorship →" + }, + { + "span_id": "span-ef87f51d1d878139ad81138f", + "text": "04 The bundles" + }, + { + "span_id": "span-aee00bc46ef98fb6dd848fb4", + "text": "Curated course sequences at a lower combined price, matched to where you're starting from." + }, + { + "span_id": "span-06c7887bd23193f7647d0eac", + "text": "Bundle · 4 learning products" + }, + { + "span_id": "span-b452bb75121b9a7584535910", + "text": "Start from zero." + }, + { + "span_id": "span-8c667fb482d23909dfe536af", + "text": "Beginner Python, the 10-Hour LLM Fundamentals video course, Full Stack AI Engineering, and the Building LLMs for Production e-book." + }, + { + "span_id": "span-3f1ad3dc9b66475dc4216820", + "text": "$727 $599 Save $128 View bundle →" + } + ] + }, + { + "chunk_id": "tai-c550fb85d206b50579025cb4", + "text": "From Developer to Advanced AI Engineer\nAlready code? The 10-Hour LLM Fundamentals crash course, Full Stack AI Engineering, Agent Engineering, and the e-book: the road to senior AI capability. $1,078 $948 Save $130 View bundle → Bundle · 6 learning products\n\nGet It All: From Novice to Expert\nThe complete curriculum for anyone: every course above plus the e-book, from your first line of Python to certified agent engineering. $1,575 $1,349 Save $226 View bundle →\n\nNot sure where to start?\nAnswer the two questions above for our recommendation, and our team reads every message. Find your path Talk to us Prices in USD · Students get 50% off all courses", + "heading": "From Developer to Advanced AI Engineer", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-df504f645c1d33af5817a241", + "text": "Already code?" + }, + { + "span_id": "span-b9002617189632801c69ecfc", + "text": "The 10-Hour LLM Fundamentals crash course, Full Stack AI Engineering, Agent Engineering, and the e-book: the road to senior AI capability." + }, + { + "span_id": "span-8b0434b891374a08f701cfb0", + "text": "$1,078 $948 Save $130 View bundle →" + }, + { + "span_id": "span-d7b97aab496ccd8b5dbaf891", + "text": "Bundle · 6 learning products" + }, + { + "span_id": "span-ba62cea6c0ac3ea23659d700", + "text": "The complete curriculum for anyone: every course above plus the e-book, from your first line of Python to certified agent engineering." + }, + { + "span_id": "span-6e67bc981e7c28662cf797f9", + "text": "$1,575 $1,349 Save $226 View bundle →" + }, + { + "span_id": "span-0ab9b58c637b632ec37d1a5e", + "text": "Answer the two questions above for our recommendation, and our team reads every message." + }, + { + "span_id": "span-afae5c6d76983559ba711844", + "text": "Find your path Talk to us" + }, + { + "span_id": "span-6b5dc7ee0a40882190ab4819", + "text": "Prices in USD · Students get 50% off all courses" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "a3a2886ba0f28a94629310ac731190d3f51de1fb3b9364237e223f5cadb7af77", + "content_hash": "a3a2886ba0f28a94629310ac731190d3f51de1fb3b9364237e223f5cadb7af77", + "authority": "official_site", + "status": "excluded", + "retrieval_eligible": false, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/bundles/" + ], + "excluded_reason": "catalog summary can conflict with canonical bundle detail pages", + "evidence_hash": "932282ff940559366ef2face68ca03808f25a3ecbd59d50bab0547e3bb0e3dc8" + }, + { + "discovered_url": "https://towardsai.com/academy/mentorship/", + "url": "https://towardsai.com/academy/mentorship/", + "canonical_url": "https://towardsai.com/academy/mentorship/", + "host": "towardsai.com", + "path": "/academy/mentorship", + "kind": "mentorship", + "offer_id": "mentorship", + "entity_id": "offer:mentorship", + "title": "Towards AI Mentorship · Senior AI Engineers On Call for Your Code and Career", + "review_title": "Mentorship", + "reviewed_summary": "Mentorship includes one course, the 10-Hour LLM Fundamentals video course from day one; the other currently listed courses are 25% off.", + "meta_description": "A subscription that puts 15+ senior AI engineers behind your career: live mentoring twice a week, a monthly production blueprint, async technical and career answers, and resume reviews in 48-72h. $899/year (about $75/month) or $99/month.", + "headings": [ + "Stuck on your AI build? Every decision, verified by a senior engineer", + "Where do your hard questions go today?", + "Build like the top 1% of AI talent, for less than a gym membership", + "A senior team on tap", + "Assets that compound", + "A price that stays small", + "The mentorship, explained", + "Does one of these sound like you?", + "The only AI person at the company", + "Building solo or freelance", + "Engineer moving into AI", + "Courses done, career not started", + "Everything inside the mentorship", + "Any question, an engineer's answer: live or in writing", + "Live Q&A, twice a week", + "Async help, any day", + "Proof, feedback, and people who hire", + "Expert reviews", + "Industry workshops", + "A production blueprint", + "The curriculum comes with the team", + "Meet the engineers and mentors on call", + "Louis-François Bouchard", + "Louie Peters", + "Samridhi Vaid", + "Omar Solano", + "Fabio Chiusano", + "Rucha Bhide", + "Why AI leaders and students trust this team", + "What students say about learning with this team", + "What #ask-an-engineer looks like", + "One price. Everything.", + "Frequently asked questions", + "Your next hard question deserves better than a chatbot's guess" + ], + "text": "Towards AI Mentorship: AI Consultancy Made Accessible to Individuals\n\nStuck on your AI build? Every decision, verified by a senior engineer\nSkill gets you to a demo; shipping and getting hired take senior judgment. This subscription puts a senior team behind you: engineers on call every week, new production assets handed over every month. Compare your options ↓ Louis-François Bouchard (Co-founder & CTO), Louie Peters (Co-founder & CEO), and the senior team on call. Meet them ↓ 15+ engineers and mentors · 10,000+ developers trained · 21+ production systems shipped The subscription Yearly · save 24% Monthly $1,188 $899 /year About $75 a month, billed once a year. Weekly Two live Q&A calls with senior engineers Any day Async answers in writing, whatever your timezone Monthly A production blueprint handed over, plus a guest workshop Included Expert reviews in 48–72h and the $199 foundational course Join the Mentorship 30-day money-back guarantee The engineers behind this mentorship have built AI systems for and have taught 500,000+ AI practitioners since 2019 01 The comparison\n\nWhere do your hard questions go today?\nChatbots hallucinate, communities answer on their own schedule, and senior consultants price by the hour. ~$75 /month · billed yearly ($899) Save 24% vs $99 month to month. Everything included 15+ senior engineers on call Answers you can trust, live Live Q&A twice a week Written answers, any day Resume & project reviews in 48–72h A production blueprint every month Courses & workshops Swipe to compare → ✓ Included ~ Limited or billed extra ✗ Not offered Get the senior team → $899/year See everything included ↓ 02 The value\n\nBuild like the top 1% of AI talent, for less than a gym membership\nThe top 1% have senior engineers catching their mistakes while they're still cheap to fix. From $75 a month, that team is behind you, and everything it delivers stacks.\n\nA senior team on tap\n$200–500/hour elsewhere Consultant-grade judgment: live twice a week, in writing any day.\n\nAssets that compound\nNew every month Every month: a production blueprint, a guest workshop, expert reviews. The $199 foundational course from day one. Nothing resets.\n\nA price that stays small\nFrom $75/month Bought separately: $2,000+ a month. Here: $75 a month billed yearly, or $99 monthly. Senior AI expertise was an enterprise line item. Now it's from $75 a month . Start compounding → $899/year Meet the engineers behind it ↓ Watch\n\nThe mentorship, explained\n03 For you\n\nDoes one of these sound like you?\nFour situations, one fix. If you recognise yours, the senior team slots straight in.\n\nThe only AI person at the company\nYou make architecture calls nobody checks. Now a senior team does: before the sprint, not after the incident.\n\nBuilding solo or freelance\nClients assume there's a team behind you. Now there is: review, patterns, a sanity-check before you sign.\n\nEngineer moving into AI\nYou have the software depth, not the AI judgment. The blueprints supply the judgment; the reviews build the proof.\n\nCourses done, career not started\nYou can't get experience without a job. The builds are the experience: shipped systems, expert feedback, a resume the hiring side reads. Why solo builders stall Resumes Screened for shipped production work. Certificates get filtered out. Demos ~95% of enterprise GenAI pilots never reach production ROI (MIT, 2025). Production Evals, monitoring, cost, reliability: everything between demo and deployed. Self-review Developers felt 20% faster with AI. Measured: 19% slower (METR, 2025). One fix for all four: senior review. 04 The offer\n\nEverything inside the mentorship\nTwo jobs: ship the project in front of you, land the role after it. Everything inside maps to one of them. 8+ Live calls twice a week Any day Written answers whatever your timezone 1× Guest workshop monthly, recorded 1× Production blueprint monthly, handed over For the project on your desk\n\nAny question, an engineer's answer: live or in writing\n\nLive Q&A, twice a week\n30 minutes per call An open call with a senior engineer: an architecture decision, a bug, a career move. Ask, or just listen.\n\nAsync help, any day\nAnswered in writing Post in the members' channel; an engineer answers in writing, whatever your timezone. See a real thread ↓ “Should I fine-tune, or is this a retrieval problem?” · “My agent passes tests and breaks with real users, where do I start?” · “Two offers, which stack matters more in three years?” For your job search and your growth\n\nProof, feedback, and people who hire\n\nExpert reviews\nWritten, in 48–72h Resume, LinkedIn, portfolio, or a project: written feedback from people who know what the hiring side discards. Resubmit as your work evolves.\n\nIndustry workshops\nMonthly · recorded Live sessions with engineers from companies like Google and OpenAI, and with senior AI engineers in our network like Paul Iusztin.\n\nA production blueprint\nMonthly · handed over Best practices we've learned, or a full system we've built for clients, handed over with the judgment calls behind it. First up: the exact setup our own engineers use to work with agents. Member bonuses\n\nThe curriculum comes with the team\nWhat Full price As a member 10-Hour LLM Fundamentals video course Five in-depth 2-hour video sessions, from a basic prompt to a full production rollout $199 Included from day one Full Stack AI Engineering · Agent Engineering · Master AI for Work $349–499 each 25% off, always New courses, while we build them Full price at launch Alpha access at 50–90% off, in exchange for the feedback that shapes the course. Agent Engineering was built this way. Earned pathways Standout members earn internships with our team and publication to Towards AI's 500,000+ readers: a public credential with your name on it. Earned, never promised. How pathways are earned + Standout means consistent shipped work: blueprint builds finished and shared, thoughtful answers in the community, write-ups that hold up to review. When we see it, we reach out. The pathways: an internship with our engineering team, articles published under your own name in the Towards AI publication, and paid course development with us. Never guaranteed, always earned. 05 Your team\n\nMeet the engineers and mentors on call\n\nLouis-François Bouchard\nCo-founder & CTO · RAG & agents Ex-PhD at Mila · co-author of Building LLMs for Production · keynote speaker, AI Engineer World's Fair\n\nLouie Peters\nCo-founder & CEO · Industry & career strategy Ex-J.P. Morgan VP · leads Towards AI's enterprise AI work and industry newsletter · hiring engineers and advising the companies that hire them\n\nSamridhi Vaid\nML Engineer · Multimodal LLMs & evaluation Built an AI scribe used by 200+ physicians across 105 facilities · agent-based eval pipelines, deployed on GCP & AWS\n\nOmar Solano\nAI Solutions Engineer · LLM applications Builds Towards AI's production LLM systems · workshop instructor, AI Engineer World's Fair\n\nFabio Chiusano\nAI Engineer · NLP & retrieval systems MSc, Politecnico di Milano · previously Machine Learning Reply · writes one of the most-read NLP newsletters in the field\n\nRucha Bhide\nMentorship Program Manager · Portfolio & communication Runs the program day to day · reviews documentation and project write-ups so strong work reads as strong Plus Jaiganesan N (AI Engineer, RAG & agents) and 8 more engineers, mentors, and reviewers. 06 Trust\n\nWhy AI leaders and students trust this team\nThe same team these leaders endorse for AI education is on call inside the mentorship. Student reviews\n\nWhat students say about learning with this team\n07 Inside the channel\n\nWhat #ask-an-engineer looks like\nOur engineers answer in writing, any day of the week. A member's RAG evals score high while real users complain; Fabio catches why. Every question gets an engineer's answer. The next one could be yours. Start asking → $899/year 08 Pricing\n\nOne price. Everything.\nEvery member gets the full mentorship. The only choice is how you pay. Yearly Save 24% · $289 off $1,188 $899 /year About $75 a month, billed once a year. Twelve months for the price of nine. 30-day money-back guarantee. Start yearly → $899/year Monthly $99 /month $1,188 over a year. The full mentorship, month to month. Cancel anytime. Start monthly → $99/month 09 Questions\n\nFrequently asked questions\nWhat if I don't have questions every month? + The core arrives whether you ask or not: eight live Q&A calls, a new production blueprint, and a workshop every month. The async channel is there for the day the wall appears. How is this different from a $150/month mentor? + A mentor is one person with one specialty. This is a team of 15+ engineers and mentors: live Q&A twice a week, a monthly production blueprint, workshops, and expert reviews. One subscription, a different specialist for each kind of question. What if I can't make the live calls? + Two run every week, the async channel answers any day in any timezone, and workshops are recorded. Can I cancel anytime? + Yes. The monthly mentorship can be cancelled at any time. The yearly plan carries a 30-day money-back guarantee. How does yearly billing work? + You pay $899 once a year, which works out to about $75 a month, 24% less than paying month to month. It comes with a 30-day money-back guarantee. When do live sessions run? + Twice a week at rotating times, so a call lands in your waking hours wherever you are. Miss one and the next is days away, with the async channel in between. What happens to my course access if I cancel? + Course access is active while your mentorship is active. If you cancel, you can upgrade the courses you've been taking to lifetime access at a reduced price. Where does the blueprint code live? + Handed over as working material: code, prompts, configs, and the practices around them. What you build on top is yours, in your own GitHub. How do the internship and publication pathways work? + Earned by strong work in the monthly builds and community: internships with our team, publishing in the Towards AI publication, course development. Never guaranteed. How is this different from buying a course? + Courses teach; the mentorship keeps you building with a team behind you. The $199 foundational course is included, but the core is live Q&A twice a week, the async channel, the monthly blueprint, and expert reviews, all compounding the longer you stay. Still deciding? Ask us whether the mentorship fits your situation, and we will answer honestly. Ask us directly → Yearly $899, about $75/month · Monthly $99\n\nYour next hard question deserves better than a chatbot's guess\nEvery question answered by a real engineer. A new production blueprint handed over each month. Get senior AI expertise → $899/year Or $99/month, cancel anytime Cancel anytime · 30-day money-back on yearly", + "links": [ + { + "text": "Join the Mentorship", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "Get the senior team → $899/year", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "Start compounding → $899/year", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "10-Hour LLM Fundamentals video course", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "Start asking → $899/year", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "Start yearly → $899/year", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "Start monthly → $99/month", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4565072" + }, + { + "text": "Ask us directly →", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "Get senior AI expertise → $899/year", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4748108" + }, + { + "text": "Or $99/month, cancel anytime", + "url": "https://academy.towardsai.net/cart/add_product/3625904?price_id=4565072" + } + ], + "chunks": [ + { + "chunk_id": "tai-0b9e3f6f8ba31ffed38a7d49", + "text": "Towards AI Mentorship: AI Consultancy Made Accessible to Individuals\n\nStuck on your AI build? Every decision, verified by a senior engineer\nSkill gets you to a demo; shipping and getting hired take senior judgment. This subscription puts a senior team behind you: engineers on call every week, new production assets handed over every month. Compare your options ↓ Louis-François Bouchard (Co-founder & CTO), Louie Peters (Co-founder & CEO), and the senior team on call. Meet them ↓ 15+ engineers and mentors · 10,000+ developers trained · 21+ production systems shipped The subscription Yearly · save 24% Monthly $1,188 $899 /year About $75 a month, billed once a year. Weekly Two live Q&A calls with senior engineers Any day Async answers in writing, whatever your timezone Monthly A production blueprint handed over, plus a guest workshop Included Expert reviews in 48–72h and the $199 foundational course Join the Mentorship 30-day money-back guarantee The engineers behind this mentorship have built AI systems for and have taught 500,000+ AI practitioners since 2019 01 The comparison\n\nWhere do your hard questions go today?\nChatbots hallucinate, communities answer on their own schedule, and senior consultants price by the hour. ~$75 /month · billed yearly ($899) Save 24% vs $99 month to month. Everything included 15+ senior engineers on call Answers you can trust, live Live Q&A twice a week Written answers, any day Resume & project reviews in 48–72h A production blueprint every month Courses & workshops Swipe to compare → ✓ Included ~ Limited or billed extra ✗ Not offered Get the senior team → $899/year See everything included ↓ 02 The value", + "heading": "Stuck on your AI build? Every decision, verified by a senior engineer", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-0881a5e2b5215ba3cb4a2b5e", + "text": "Towards AI Mentorship: AI Consultancy Made Accessible to Individuals" + }, + { + "span_id": "span-0dfa1fba3a60102ee9149f38", + "text": "Skill gets you to a demo; shipping and getting hired take senior judgment." + }, + { + "span_id": "span-3c98abfcf407f20900344acb", + "text": "This subscription puts a senior team behind you: engineers on call every week, new production assets handed over every month." + }, + { + "span_id": "span-833f2e13a17de4e155446b2e", + "text": "Compare your options ↓" + }, + { + "span_id": "span-abe7b445a73fa3396925cedf", + "text": "Louis-François Bouchard (Co-founder & CTO), Louie Peters (Co-founder & CEO), and the senior team on call." + }, + { + "span_id": "span-580f136046c0be24329869d4", + "text": "Meet them ↓" + }, + { + "span_id": "span-88af07e39d946e8008d133af", + "text": "15+ engineers and mentors · 10,000+ developers trained · 21+ production systems shipped" + }, + { + "span_id": "span-33185956bd088073238c0d70", + "text": "The subscription" + }, + { + "span_id": "span-104cec8cde5e67c5b880ade3", + "text": "Yearly · save 24% Monthly" + }, + { + "span_id": "span-de0674326fbab5393c5a2ccc", + "text": "$1,188 $899 /year" + }, + { + "span_id": "span-3ff193b0464b5873f14e1e44", + "text": "About $75 a month, billed once a year." + }, + { + "span_id": "span-da0482d341b8e881dd7ef36d", + "text": "Two live Q&A calls with senior engineers" + }, + { + "span_id": "span-b5cde45b9fb7a4a66431504a", + "text": "Any day" + }, + { + "span_id": "span-1e0e216ded6bba235d0191d4", + "text": "Async answers in writing, whatever your timezone" + }, + { + "span_id": "span-30dc555345db4eda16fd49c2", + "text": "A production blueprint handed over, plus a guest workshop" + }, + { + "span_id": "span-cb686626353ef98d04168a97", + "text": "Expert reviews in 48–72h and the $199 foundational course" + }, + { + "span_id": "span-62cf55f73fce84ce4c9c3827", + "text": "Join the Mentorship 30-day money-back guarantee" + }, + { + "span_id": "span-6ef78944b44d71d06d8c927a", + "text": "The engineers behind this mentorship have built AI systems for" + }, + { + "span_id": "span-36dbd568cc11b13f0318cade", + "text": "and have taught 500,000+ AI practitioners since 2019" + }, + { + "span_id": "span-642c7ecb1e23e201420ecae0", + "text": "01 The comparison" + }, + { + "span_id": "span-d432d9663874daaf3a31f804", + "text": "Chatbots hallucinate, communities answer on their own schedule, and senior consultants price by the hour." + }, + { + "span_id": "span-ad9906c8625d51b4425ad9dd", + "text": "~$75 /month · billed yearly ($899)" + }, + { + "span_id": "span-5ee04ed0daaf47a0a55052de", + "text": "Save 24% vs $99 month to month." + }, + { + "span_id": "span-d0a41681780600033c8a8012", + "text": "Everything included 15+ senior engineers on call" + }, + { + "span_id": "span-507cdf0d96acee60a0ce0368", + "text": "Answers you can trust, live" + }, + { + "span_id": "span-45ceb13d407858ed2c50b740", + "text": "Live Q&A twice a week" + }, + { + "span_id": "span-f3803b01c8644bdc18c8af84", + "text": "Written answers, any day" + }, + { + "span_id": "span-7c714a8e7f8a9e1fd3d37d66", + "text": "Resume & project reviews in 48–72h" + }, + { + "span_id": "span-217a82a8fcb27919e11248d4", + "text": "A production blueprint every month" + }, + { + "span_id": "span-e2b112d5b9b8f0aa78e5b660", + "text": "Courses & workshops" + }, + { + "span_id": "span-72f97e3e37bd73c0e5378d34", + "text": "Swipe to compare →" + }, + { + "span_id": "span-4f6589cfd88ccad8a8450909", + "text": "Get the senior team → $899/year See everything included ↓" + }, + { + "span_id": "span-3f00c9cbb59efbf08416c60a", + "text": "02 The value" + } + ] + }, + { + "chunk_id": "tai-f659bb5cb4b830c95cd8a69e", + "text": "Build like the top 1% of AI talent, for less than a gym membership\nThe top 1% have senior engineers catching their mistakes while they're still cheap to fix. From $75 a month, that team is behind you, and everything it delivers stacks.\n\nA senior team on tap\n$200–500/hour elsewhere Consultant-grade judgment: live twice a week, in writing any day.\n\nAssets that compound\nNew every month Every month: a production blueprint, a guest workshop, expert reviews. The $199 foundational course from day one. Nothing resets.\n\nA price that stays small\nFrom $75/month Bought separately: $2,000+ a month. Here: $75 a month billed yearly, or $99 monthly. Senior AI expertise was an enterprise line item. Now it's from $75 a month . Start compounding → $899/year Meet the engineers behind it ↓ Watch\n\nThe mentorship, explained\n03 For you\n\nDoes one of these sound like you?\nFour situations, one fix. If you recognise yours, the senior team slots straight in.\n\nThe only AI person at the company\nYou make architecture calls nobody checks. Now a senior team does: before the sprint, not after the incident.\n\nBuilding solo or freelance\nClients assume there's a team behind you. Now there is: review, patterns, a sanity-check before you sign.\n\nEngineer moving into AI\nYou have the software depth, not the AI judgment. The blueprints supply the judgment; the reviews build the proof.", + "heading": "A price that stays small", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-e458300c762a560ccbdc9178", + "text": "Build like the top 1% of AI talent, for less than a gym membership" + }, + { + "span_id": "span-42f6fdcd6faeec7d719cc95a", + "text": "The top 1% have senior engineers catching their mistakes while they're still cheap to fix." + }, + { + "span_id": "span-2aa40d3c2f44c9bd0822ca85", + "text": "From $75 a month, that team is behind you, and everything it delivers stacks." + }, + { + "span_id": "span-35deaf22eeee124d733561b0", + "text": "$200–500/hour elsewhere" + }, + { + "span_id": "span-fd10ef7bd5069626101e3594", + "text": "Consultant-grade judgment: live twice a week, in writing any day." + }, + { + "span_id": "span-d2407c8792f5dac0a55a5025", + "text": "New every month" + }, + { + "span_id": "span-33259fb8cc9353fa90a5ad3b", + "text": "Every month: a production blueprint, a guest workshop, expert reviews." + }, + { + "span_id": "span-cf765a07909d5db54f64a0f7", + "text": "The $199 foundational course from day one." + }, + { + "span_id": "span-35ca44d391d674ac062b039d", + "text": "Nothing resets." + }, + { + "span_id": "span-bd9fbfc87f28dbe8ac4d8604", + "text": "From $75/month" + }, + { + "span_id": "span-1a6101d2903a6ea22de5d1f2", + "text": "Bought separately: $2,000+ a month." + }, + { + "span_id": "span-1cafd359f03e9076b140684f", + "text": "Here: $75 a month billed yearly, or $99 monthly." + }, + { + "span_id": "span-7cb73e20f91bdec5b100a200", + "text": "Senior AI expertise was an enterprise line item." + }, + { + "span_id": "span-3754f3322e4215d98173d528", + "text": "Now it's from $75 a month ." + }, + { + "span_id": "span-c8868a1fdd462012a273272f", + "text": "Start compounding → $899/year Meet the engineers behind it ↓" + }, + { + "span_id": "span-ede182db5feab61c28a4ed5d", + "text": "03 For you" + }, + { + "span_id": "span-954ecab3d0e26de80cb545e1", + "text": "Four situations, one fix." + }, + { + "span_id": "span-ef2a33f101be49b28d2192b7", + "text": "If you recognise yours, the senior team slots straight in." + }, + { + "span_id": "span-5f4b40f405fc749b5a76e754", + "text": "You make architecture calls nobody checks." + }, + { + "span_id": "span-0cfea00e699590f0609b052b", + "text": "Now a senior team does: before the sprint, not after the incident." + }, + { + "span_id": "span-6e7d63c3064a826b215857f3", + "text": "Clients assume there's a team behind you." + }, + { + "span_id": "span-66cf90062c20e20ae40e0fa7", + "text": "Now there is: review, patterns, a sanity-check before you sign." + }, + { + "span_id": "span-cd394a86b9b505d8548c97a5", + "text": "You have the software depth, not the AI judgment." + }, + { + "span_id": "span-1aef9338ad0a48132dfb13a0", + "text": "The blueprints supply the judgment; the reviews build the proof." + } + ] + }, + { + "chunk_id": "tai-b103bd52b572b1f70d58483a", + "text": "Courses done, career not started\nYou can't get experience without a job. The builds are the experience: shipped systems, expert feedback, a resume the hiring side reads. Why solo builders stall Resumes Screened for shipped production work. Certificates get filtered out. Demos ~95% of enterprise GenAI pilots never reach production ROI (MIT, 2025). Production Evals, monitoring, cost, reliability: everything between demo and deployed. Self-review Developers felt 20% faster with AI. Measured: 19% slower (METR, 2025). One fix for all four: senior review. 04 The offer\n\nEverything inside the mentorship\nTwo jobs: ship the project in front of you, land the role after it. Everything inside maps to one of them. 8+ Live calls twice a week Any day Written answers whatever your timezone 1× Guest workshop monthly, recorded 1× Production blueprint monthly, handed over For the project on your desk\n\nAny question, an engineer's answer: live or in writing\n\nLive Q&A, twice a week\n30 minutes per call An open call with a senior engineer: an architecture decision, a bug, a career move. Ask, or just listen.\n\nAsync help, any day\nAnswered in writing Post in the members' channel; an engineer answers in writing, whatever your timezone. See a real thread ↓ “Should I fine-tune, or is this a retrieval problem?” · “My agent passes tests and breaks with real users, where do I start?” · “Two offers, which stack matters more in three years?” For your job search and your growth\n\nProof, feedback, and people who hire\n\nExpert reviews\nWritten, in 48–72h Resume, LinkedIn, portfolio, or a project: written feedback from people who know what the hiring side discards. Resubmit as your work evolves.", + "heading": "Courses done, career not started", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-a7cd47bb93ceb88c881f2e39", + "text": "You can't get experience without a job." + }, + { + "span_id": "span-b7871a1817601ba0df282179", + "text": "The builds are the experience: shipped systems, expert feedback, a resume the hiring side reads." + }, + { + "span_id": "span-05ecf34799f05a4a1be28493", + "text": "Why solo builders stall" + }, + { + "span_id": "span-f19e309470076b8ed0e43976", + "text": "Screened for shipped production work." + }, + { + "span_id": "span-0fd6f1de87e145a9834575f1", + "text": "Certificates get filtered out." + }, + { + "span_id": "span-c7b0b243b08b624cab7b8c13", + "text": "~95% of enterprise GenAI pilots never reach production ROI (MIT, 2025)." + }, + { + "span_id": "span-7c26a0ac104a01b94bc6b5c4", + "text": "Evals, monitoring, cost, reliability: everything between demo and deployed." + }, + { + "span_id": "span-fe8f5a5063f2f29bb8e2921e", + "text": "Self-review" + }, + { + "span_id": "span-f502f30792374cb9c1a07585", + "text": "Developers felt 20% faster with AI." + }, + { + "span_id": "span-d9c966ce7875f9f401dbf594", + "text": "Measured: 19% slower (METR, 2025)." + }, + { + "span_id": "span-8f2b47367362fce720d25de9", + "text": "One fix for all four: senior review." + }, + { + "span_id": "span-42f3358151211f63940fbe2b", + "text": "04 The offer" + }, + { + "span_id": "span-d36275cfbb03a246ee5cacbb", + "text": "Two jobs: ship the project in front of you, land the role after it." + }, + { + "span_id": "span-dc70bb7a75d1fff7a9df439e", + "text": "Everything inside maps to one of them." + }, + { + "span_id": "span-d34f9d7991c9308f645513ac", + "text": "8+ Live calls twice a week" + }, + { + "span_id": "span-2a8984def944ba860c15e7e3", + "text": "Any day Written answers whatever your timezone" + }, + { + "span_id": "span-3796cf23c49c4962fb1cc380", + "text": "1× Guest workshop monthly, recorded" + }, + { + "span_id": "span-3330d930a8f91eb22b3cf083", + "text": "1× Production blueprint monthly, handed over" + }, + { + "span_id": "span-cb57080a2fe09d482fb49027", + "text": "For the project on your desk" + }, + { + "span_id": "span-4415e000c9026e5c34675ac8", + "text": "Any question, an engineer's answer: live or in writing" + }, + { + "span_id": "span-fad0828c7526939debb2d9dd", + "text": "30 minutes per call" + }, + { + "span_id": "span-5437effd8346ac27fb0eea80", + "text": "An open call with a senior engineer: an architecture decision, a bug, a career move." + }, + { + "span_id": "span-b0be3142fb6c3f8ab829385c", + "text": "Ask, or just listen." + }, + { + "span_id": "span-16a04cf2e47ae1f1883f9c8f", + "text": "Answered in writing" + }, + { + "span_id": "span-8991d4bd8967f723586f57e8", + "text": "Post in the members' channel; an engineer answers in writing, whatever your timezone." + }, + { + "span_id": "span-f90cf6223ba288aeb4ecae1f", + "text": "See a real thread ↓" + }, + { + "span_id": "span-ac1b026c0e5f66745c2dc4aa", + "text": "“Should I fine-tune, or is this a retrieval problem?” · “My agent passes tests and breaks with real users, where do I start?” · “Two offers, which stack matters more in three years?”" + }, + { + "span_id": "span-de82547a3a53ac1a7aef6715", + "text": "For your job search and your growth" + }, + { + "span_id": "span-953e2c5a31c77113ff7a4949", + "text": "Proof, feedback, and people who hire" + }, + { + "span_id": "span-fe39deed7a65b826f3f92751", + "text": "Written, in 48–72h" + }, + { + "span_id": "span-52d8117ba468c9782356b2d4", + "text": "Resume, LinkedIn, portfolio, or a project: written feedback from people who know what the hiring side discards." + }, + { + "span_id": "span-2c7d3d95e75c9eddaa570891", + "text": "Resubmit as your work evolves." + } + ] + }, + { + "chunk_id": "tai-77520b6682abadc7f0f2c279", + "text": "Industry workshops\nMonthly · recorded Live sessions with engineers from companies like Google and OpenAI, and with senior AI engineers in our network like Paul Iusztin.\n\nA production blueprint\nMonthly · handed over Best practices we've learned, or a full system we've built for clients, handed over with the judgment calls behind it. First up: the exact setup our own engineers use to work with agents. Member bonuses\n\nThe curriculum comes with the team\nWhat Full price As a member 10-Hour LLM Fundamentals video course Five in-depth 2-hour video sessions, from a basic prompt to a full production rollout $199 Included from day one Full Stack AI Engineering · Agent Engineering · Master AI for Work $349–499 each 25% off, always New courses, while we build them Full price at launch Alpha access at 50–90% off, in exchange for the feedback that shapes the course. Agent Engineering was built this way. Earned pathways Standout members earn internships with our team and publication to Towards AI's 500,000+ readers: a public credential with your name on it. Earned, never promised. How pathways are earned + Standout means consistent shipped work: blueprint builds finished and shared, thoughtful answers in the community, write-ups that hold up to review. When we see it, we reach out. The pathways: an internship with our engineering team, articles published under your own name in the Towards AI publication, and paid course development with us. Never guaranteed, always earned. 05 Your team\n\nMeet the engineers and mentors on call\n\nLouis-François Bouchard\nCo-founder & CTO · RAG & agents Ex-PhD at Mila · co-author of Building LLMs for Production · keynote speaker, AI Engineer World's Fair", + "heading": "The curriculum comes with the team", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-ec40d5cf1e0d0bebe081ffb6", + "text": "Monthly · recorded" + }, + { + "span_id": "span-e65a2467d5e5a5a9ef02dee3", + "text": "Live sessions with engineers from companies like Google and OpenAI, and with senior AI engineers in our network like Paul Iusztin." + }, + { + "span_id": "span-f32aebbd7df4635bccadba4b", + "text": "Monthly · handed over" + }, + { + "span_id": "span-49e3814c76298af919d90a5f", + "text": "Best practices we've learned, or a full system we've built for clients, handed over with the judgment calls behind it." + }, + { + "span_id": "span-5c73f1a32a7a32208f0e911c", + "text": "First up: the exact setup our own engineers use to work with agents." + }, + { + "span_id": "span-8b7aa7f91f52f8bae80c6ae7", + "text": "Member bonuses" + }, + { + "span_id": "span-39763323a43279c1ee15192c", + "text": "10-Hour LLM Fundamentals video course Five in-depth 2-hour video sessions, from a basic prompt to a full production rollout $199 Included from day one" + }, + { + "span_id": "span-b78c3bde1f0aeea5b9073e13", + "text": "Full Stack AI Engineering · Agent Engineering · Master AI for Work $349–499 each 25% off, always" + }, + { + "span_id": "span-366d9eea7c6111181476f4d6", + "text": "New courses, while we build them Full price at launch Alpha access at 50–90% off, in exchange for the feedback that shapes the course." + }, + { + "span_id": "span-1be14983a1a09cc53b30c442", + "text": "Agent Engineering was built this way." + }, + { + "span_id": "span-3b415d84c9dcec6ca81647b6", + "text": "Earned pathways Standout members earn internships with our team and publication to Towards AI's 500,000+ readers: a public credential with your name on it." + }, + { + "span_id": "span-779b91f931037d22734a7bf5", + "text": "Earned, never promised." + }, + { + "span_id": "span-a7c8451e3bbb980ffd86a358", + "text": "How pathways are earned + Standout means consistent shipped work: blueprint builds finished and shared, thoughtful answers in the community, write-ups that hold up to review." + }, + { + "span_id": "span-fd43be17e78ffcf7b0360f6f", + "text": "When we see it, we reach out." + }, + { + "span_id": "span-c006d5492206b8978e65ec40", + "text": "The pathways: an internship with our engineering team, articles published under your own name in the Towards AI publication, and paid course development with us." + }, + { + "span_id": "span-f91523a932e7379b3c896c67", + "text": "Never guaranteed, always earned." + }, + { + "span_id": "span-50fb3425239d30a8cda1f9e7", + "text": "05 Your team" + }, + { + "span_id": "span-60c75958d0bc76db55249e6c", + "text": "Meet the engineers and mentors on call" + }, + { + "span_id": "span-7d996539ec19f74334cbe74e", + "text": "Co-founder & CTO · RAG & agents Ex-PhD at Mila · co-author of Building LLMs for Production · keynote speaker, AI Engineer World's Fair" + } + ] + }, + { + "chunk_id": "tai-11fa8a59d901a4a7b7f636c0", + "text": "Louie Peters\nCo-founder & CEO · Industry & career strategy Ex-J.P. Morgan VP · leads Towards AI's enterprise AI work and industry newsletter · hiring engineers and advising the companies that hire them\n\nSamridhi Vaid\nML Engineer · Multimodal LLMs & evaluation Built an AI scribe used by 200+ physicians across 105 facilities · agent-based eval pipelines, deployed on GCP & AWS\n\nOmar Solano\nAI Solutions Engineer · LLM applications Builds Towards AI's production LLM systems · workshop instructor, AI Engineer World's Fair\n\nFabio Chiusano\nAI Engineer · NLP & retrieval systems MSc, Politecnico di Milano · previously Machine Learning Reply · writes one of the most-read NLP newsletters in the field\n\nRucha Bhide\nMentorship Program Manager · Portfolio & communication Runs the program day to day · reviews documentation and project write-ups so strong work reads as strong Plus Jaiganesan N (AI Engineer, RAG & agents) and 8 more engineers, mentors, and reviewers. 06 Trust\n\nWhy AI leaders and students trust this team\nThe same team these leaders endorse for AI education is on call inside the mentorship. Student reviews\n\nWhat students say about learning with this team\n07 Inside the channel\n\nWhat #ask-an-engineer looks like\nOur engineers answer in writing, any day of the week. A member's RAG evals score high while real users complain; Fabio catches why. Every question gets an engineer's answer. The next one could be yours. Start asking → $899/year 08 Pricing", + "heading": "Rucha Bhide", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-99e29f60bf371db6a7db3021", + "text": "Co-founder & CEO · Industry & career strategy Ex-J.P." + }, + { + "span_id": "span-9e738e9958433092173ae4ce", + "text": "Morgan VP · leads Towards AI's enterprise AI work and industry newsletter · hiring engineers and advising the companies that hire them" + }, + { + "span_id": "span-3256d44ab947471ceef88be6", + "text": "ML Engineer · Multimodal LLMs & evaluation Built an AI scribe used by 200+ physicians across 105 facilities · agent-based eval pipelines, deployed on GCP & AWS" + }, + { + "span_id": "span-af7541ee3408c590476d43ba", + "text": "AI Solutions Engineer · LLM applications Builds Towards AI's production LLM systems · workshop instructor, AI Engineer World's Fair" + }, + { + "span_id": "span-d51652a2d1fd87d6f05379ce", + "text": "AI Engineer · NLP & retrieval systems MSc, Politecnico di Milano · previously Machine Learning Reply · writes one of the most-read NLP newsletters in the field" + }, + { + "span_id": "span-fb0685a8f9b71ac062fd018d", + "text": "Mentorship Program Manager · Portfolio & communication Runs the program day to day · reviews documentation and project write-ups so strong work reads as strong" + }, + { + "span_id": "span-3e03c99b709cb11f5bd06e4a", + "text": "Plus Jaiganesan N (AI Engineer, RAG & agents) and 8 more engineers, mentors, and reviewers." + }, + { + "span_id": "span-f1965f81ff7742d6171c0854", + "text": "06 Trust" + }, + { + "span_id": "span-74f93db9ba063089e191339d", + "text": "The same team these leaders endorse for AI education is on call inside the mentorship." + }, + { + "span_id": "span-cf8c9dc45fdad5c205a2d92f", + "text": "Student reviews" + }, + { + "span_id": "span-7069f0be9d4d3f19ae09827f", + "text": "07 Inside the channel" + }, + { + "span_id": "span-a8eb24717669dffddf1799dc", + "text": "Our engineers answer in writing, any day of the week." + }, + { + "span_id": "span-2fe94beca7300c2f2b3e9fef", + "text": "A member's RAG evals score high while real users complain; Fabio catches why." + }, + { + "span_id": "span-ac529e29b19ecac64a2f3885", + "text": "Every question gets an engineer's answer." + }, + { + "span_id": "span-c6d4b69a50d6d77d909a1d67", + "text": "The next one could be yours." + }, + { + "span_id": "span-4f39bf3effc8c22ecbfea750", + "text": "Start asking → $899/year" + }, + { + "span_id": "span-262d02f58b9f690edeb063a3", + "text": "08 Pricing" + } + ] + }, + { + "chunk_id": "tai-6ff4a5336042a83949074c35", + "text": "One price. Everything.\nEvery member gets the full mentorship. The only choice is how you pay. Yearly Save 24% · $289 off $1,188 $899 /year About $75 a month, billed once a year. Twelve months for the price of nine. 30-day money-back guarantee. Start yearly → $899/year Monthly $99 /month $1,188 over a year. The full mentorship, month to month. Cancel anytime. Start monthly → $99/month 09 Questions", + "heading": "One price. Everything.", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-a2dadbe9eb5a4e51dac90bb7", + "text": "Every member gets the full mentorship." + }, + { + "span_id": "span-d7823d7f55a7f17ec3c3480b", + "text": "The only choice is how you pay." + }, + { + "span_id": "span-38227f4bd36c7e7835bc8ad0", + "text": "Yearly Save 24% · $289 off" + }, + { + "span_id": "span-3a3a982c2ce57ed7182f5e6c", + "text": "$1,188 $899 /year" + }, + { + "span_id": "span-306413622b794801ae31a3f6", + "text": "About $75 a month, billed once a year." + }, + { + "span_id": "span-bbc9ef4693c5096f741247c1", + "text": "Twelve months for the price of nine." + }, + { + "span_id": "span-c506ff8ecfaa8ec0935375b0", + "text": "30-day money-back guarantee." + }, + { + "span_id": "span-81698e517f90cc723db482ac", + "text": "Start yearly → $899/year" + }, + { + "span_id": "span-28e60443401abd385a4fa7fb", + "text": "$99 /month" + }, + { + "span_id": "span-428d7f08333662eeba0d2c7c", + "text": "$1,188 over a year." + }, + { + "span_id": "span-0b4e8e6838aefa1d47c16227", + "text": "The full mentorship, month to month." + }, + { + "span_id": "span-187c66ceadc1457ba7c72656", + "text": "Cancel anytime." + }, + { + "span_id": "span-2c5d070770b573badd467dee", + "text": "Start monthly → $99/month" + }, + { + "span_id": "span-3592a037bd9356e213d4789c", + "text": "09 Questions" + } + ] + }, + { + "chunk_id": "tai-39da32a2957350017b358bf1", + "text": "Frequently asked questions\nWhat if I don't have questions every month? + The core arrives whether you ask or not: eight live Q&A calls, a new production blueprint, and a workshop every month. The async channel is there for the day the wall appears. How is this different from a $150/month mentor? + A mentor is one person with one specialty. This is a team of 15+ engineers and mentors: live Q&A twice a week, a monthly production blueprint, workshops, and expert reviews. One subscription, a different specialist for each kind of question. What if I can't make the live calls? + Two run every week, the async channel answers any day in any timezone, and workshops are recorded. Can I cancel anytime? + Yes. The monthly mentorship can be cancelled at any time. The yearly plan carries a 30-day money-back guarantee. How does yearly billing work? + You pay $899 once a year, which works out to about $75 a month, 24% less than paying month to month. It comes with a 30-day money-back guarantee. When do live sessions run? + Twice a week at rotating times, so a call lands in your waking hours wherever you are. Miss one and the next is days away, with the async channel in between. What happens to my course access if I cancel? + Course access is active while your mentorship is active. If you cancel, you can upgrade the courses you've been taking to lifetime access at a reduced price. Where does the blueprint code live? + Handed over as working material: code, prompts, configs, and the practices around them. What you build on top is yours, in your own GitHub. How do the internship and publication pathways work? + Earned by strong work in the monthly builds and community: internships with our team, publishing in the Towards AI publication, course development. Never guaranteed. How is this", + "heading": "Frequently asked questions", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-0752091dda7657793fd5d7fe", + "text": "The core arrives whether you ask or not: eight live Q&A calls, a new production blueprint, and a workshop every month." + }, + { + "span_id": "span-aa4d9fa30dfe6d03af78b2d4", + "text": "The async channel is there for the day the wall appears." + }, + { + "span_id": "span-178cbe42ee9c3f566e27a95e", + "text": "A mentor is one person with one specialty." + }, + { + "span_id": "span-1cba7bbc7a3960d167d7ea4d", + "text": "This is a team of 15+ engineers and mentors: live Q&A twice a week, a monthly production blueprint, workshops, and expert reviews." + }, + { + "span_id": "span-065191e5a6cb643e657735a7", + "text": "One subscription, a different specialist for each kind of question." + }, + { + "span_id": "span-de947c6654bbc755a1b908f1", + "text": "Two run every week, the async channel answers any day in any timezone, and workshops are recorded." + }, + { + "span_id": "span-0845ec41955c7e17da92a28c", + "text": "The monthly mentorship can be cancelled at any time." + }, + { + "span_id": "span-1e7f865d7252b2ae3e82af8b", + "text": "The yearly plan carries a 30-day money-back guarantee." + }, + { + "span_id": "span-f9f7c60eb5332688bfd10d23", + "text": "You pay $899 once a year, which works out to about $75 a month, 24% less than paying month to month." + }, + { + "span_id": "span-2324038bd0bfe1c1adbf2845", + "text": "It comes with a 30-day money-back guarantee." + }, + { + "span_id": "span-da1bbf092991dff9283f4066", + "text": "Twice a week at rotating times, so a call lands in your waking hours wherever you are." + }, + { + "span_id": "span-e45f715ba26d6dd3120a62db", + "text": "Miss one and the next is days away, with the async channel in between." + }, + { + "span_id": "span-06866d26afa48a05481d6f40", + "text": "Course access is active while your mentorship is active." + }, + { + "span_id": "span-56c83fa71f37cef381060f95", + "text": "If you cancel, you can upgrade the courses you've been taking to lifetime access at a reduced price." + }, + { + "span_id": "span-a090dec0474877ffe433f0dc", + "text": "Handed over as working material: code, prompts, configs, and the practices around them." + }, + { + "span_id": "span-8a44549d5cefa0491cff041b", + "text": "What you build on top is yours, in your own GitHub." + }, + { + "span_id": "span-81deedae9cd3296472a04f6d", + "text": "Earned by strong work in the monthly builds and community: internships with our team, publishing in the Towards AI publication, course development." + }, + { + "span_id": "span-e74be9dbcb48ad9bf3d3fdfb", + "text": "Never guaranteed." + } + ] + }, + { + "chunk_id": "tai-0a9526738213e8cc755da495", + "text": "Frequently asked questions\ndifferent from buying a course? + Courses teach; the mentorship keeps you building with a team behind you. The $199 foundational course is included, but the core is live Q&A twice a week, the async channel, the monthly blueprint, and expert reviews, all compounding the longer you stay. Still deciding? Ask us whether the mentorship fits your situation, and we will answer honestly. Ask us directly → Yearly $899, about $75/month · Monthly $99\n\nYour next hard question deserves better than a chatbot's guess\nEvery question answered by a real engineer. A new production blueprint handed over each month. Get senior AI expertise → $899/year Or $99/month, cancel anytime Cancel anytime · 30-day money-back on yearly", + "heading": "Frequently asked questions", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-d1cfdec97981cd9dfbe4f411", + "text": "Courses teach; the mentorship keeps you building with a team behind you." + }, + { + "span_id": "span-02f68ebf0d66aab0e85a1212", + "text": "The $199 foundational course is included, but the core is live Q&A twice a week, the async channel, the monthly blueprint, and expert reviews, all compounding the longer you stay." + }, + { + "span_id": "span-9d346396a66cf725137656bd", + "text": "Still deciding?" + }, + { + "span_id": "span-b7efab3e61680851a19e3fee", + "text": "Ask us whether the mentorship fits your situation, and we will answer honestly." + }, + { + "span_id": "span-5bf1466eaff4fd5885ad97a9", + "text": "Ask us directly →" + }, + { + "span_id": "span-3c5c401d3e43411c92a18f29", + "text": "Yearly $899, about $75/month · Monthly $99" + }, + { + "span_id": "span-3b742103dbb8098ada28294d", + "text": "Every question answered by a real engineer." + }, + { + "span_id": "span-2090768f55429987aaee9373", + "text": "A new production blueprint handed over each month." + }, + { + "span_id": "span-8dc6bbfce4b0dfcf265c0b00", + "text": "Get senior AI expertise → $899/year Or $99/month, cancel anytime" + }, + { + "span_id": "span-821273c7555fe20020b0fd4a", + "text": "Cancel anytime · 30-day money-back on yearly" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "ca67b011d44fc15172286c9bee72075a6a0e1111c7b19c4128b3312a354bdc54", + "content_hash": "ca67b011d44fc15172286c9bee72075a6a0e1111c7b19c4128b3312a354bdc54", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/mentorship/" + ], + "evidence_hash": "e231445b52198054b6ae445eb32feba64a8bf26fe76a4067510dbab682130f27" + }, + { + "discovered_url": "https://towardsai.com/academy/about/", + "url": "https://towardsai.com/academy/about/", + "canonical_url": "https://towardsai.com/academy/about/", + "host": "towardsai.com", + "path": "/academy/about", + "kind": "page", + "offer_id": null, + "entity_id": "page:towardsai.com/academy/about", + "title": "About Us · Towards AI Academy", + "review_title": "About", + "reviewed_summary": "Towards AI Academy and team background.", + "meta_description": "Towards AI trains developers for production AI through project-based courses, in-person bootcamps, and customized enterprise programs. 10,000+ engineers enrolled since 2019.", + "headings": [ + "Training developers for production AI", + "Meet our founders", + "Three ways we train AI-first teams", + "Project-based courses", + "In-person bootcamps", + "Customized enterprise programs", + "Industry leaders on our resources", + "About Towards AI", + "Who are these resources for?", + "Find your course" + ], + "text": "About Towards AI\n\nTraining developers for production AI\nThrough project-based courses, in-person bootcamps, and customized enterprise programs. 10,000+ engineers enrolled since 2019 500k+ learners across our platforms 15+ AI practitioners on the team Including teams at 01 Leadership\n\nMeet our founders\n“We are at the beginning of the most significant technological shift in history. The solutions that define this era, in healthcare, energy, finance, and education, will be built by AI engineers. I am building Towards AI to be where those engineers are made.” Louie Peters · Co-founder & CEO “I left academia to solve real problems and discovered the hardest one was finding training that reflected how production AI actually works. So I built them. Everything we create at Towards AI is focused on one thing: you will learn to build and ship production-grade AI systems.” Louis-François Bouchard · Co-founder & CTO 02 What we do\n\nThree ways we train AI-first teams\n1 / 3 For individuals\n\nProject-based courses\nThe most in-depth path from developer to AI engineer Most developers hit the same wall: tutorials that work in notebooks but fall apart in production. Our courses are built around one project taken end-to-end, from raw data pipeline to live deployment, including the failure modes notebooks never show you. By the time you finish, you've shipped something that works. Become an AI Engineer 2 / 3 For teams\n\nIn-person bootcamps\nIntensive training to accelerate AI adoption within your teams Your developers know how to code. The gap is knowing how to build reliably on top of probabilistic systems, so LLM features hold up under real usage. Our bootcamps close that gap directly, tailored to your stack and the specific problems your team is trying to solve. Get in touch 3 / 3 For organizations\n\nCustomized enterprise programs\nEnd-to-end AI consulting for organizations building with LLMs Most enterprise teams lack the production experience to know which AI initiatives are worth building. Our team of 15+ AI practitioners works directly with yours to identify the right use cases and build systems that actually deliver. Learn more Engagement formats Advisory Strategic guidance on use cases, models, and architecture Sprints Hands-on development sprints with your team Custom build LLM systems built end-to-end, handed over documented Enablement Training that keeps the capability in-house 03 Trusted voices\n\nIndustry leaders on our resources\nSince 2019\n\nAbout Towards AI\nSince 2019, we have been bridging the gap between AI research and what it actually takes to build with it in industry. What started as a newsletter has grown into one of the largest AI learning communities: 500,000+ learners, 200,000+ newsletter subscribers, and a 100,000-member Discord where engineers share what they are actually building. Behind every course, bootcamp, and enterprise program is a team of 15+ AI practitioners who work on production systems daily. They built the training they couldn't find anywhere else. 04 Who we serve\n\nWho are these resources for?\nOur material is designed for highly motivated learners who are ready to invest in their future. We offer courses for every skill level, and none of them are passive: you'll be actively building through hands-on projects rather than sitting through long videos. Whether you're a beginner or an experienced coder, the content challenges you and grows with your skills. Expect to dedicate real time to practical assignments that reflect production applications. Our learners have moved into high-paying AI roles and led AI initiatives inside their companies by gaining the skills to build and operate advanced AI systems. Start here\n\nFind your course\nFive programs, one pathway: from beginner Python to full stack AI engineering and production agents. Browse the pathway Talk to us about team training", + "links": [ + { + "text": "Become an AI Engineer", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Get in touch", + "url": "https://towardsai.com/academy/contact/" + }, + { + "text": "Learn more", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "Browse the pathway", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Talk to us about team training", + "url": "https://towardsai.com/academy/contact/" + } + ], + "chunks": [ + { + "chunk_id": "tai-b690c6401ae7fb0e5e2d720d", + "text": "About Towards AI\n\nTraining developers for production AI\nThrough project-based courses, in-person bootcamps, and customized enterprise programs. 10,000+ engineers enrolled since 2019 500k+ learners across our platforms 15+ AI practitioners on the team Including teams at 01 Leadership\n\nMeet our founders\n“We are at the beginning of the most significant technological shift in history. The solutions that define this era, in healthcare, energy, finance, and education, will be built by AI engineers. I am building Towards AI to be where those engineers are made.” Louie Peters · Co-founder & CEO “I left academia to solve real problems and discovered the hardest one was finding training that reflected how production AI actually works. So I built them. Everything we create at Towards AI is focused on one thing: you will learn to build and ship production-grade AI systems.” Louis-François Bouchard · Co-founder & CTO 02 What we do\n\nThree ways we train AI-first teams\n1 / 3 For individuals\n\nProject-based courses\nThe most in-depth path from developer to AI engineer Most developers hit the same wall: tutorials that work in notebooks but fall apart in production. Our courses are built around one project taken end-to-end, from raw data pipeline to live deployment, including the failure modes notebooks never show you. By the time you finish, you've shipped something that works. Become an AI Engineer 2 / 3 For teams", + "heading": "Meet our founders", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-6253b2644f1710021b7c19d5", + "text": "About Towards AI" + }, + { + "span_id": "span-0d9a2ab9260f8c661a84c1d2", + "text": "Through project-based courses, in-person bootcamps, and customized enterprise programs." + }, + { + "span_id": "span-613984239ddb67d601e0f827", + "text": "10,000+ engineers enrolled since 2019" + }, + { + "span_id": "span-705bf8239b23c91dc9c85922", + "text": "500k+ learners across our platforms" + }, + { + "span_id": "span-03f2d886cee9cd53e2859b4f", + "text": "15+ AI practitioners on the team" + }, + { + "span_id": "span-fc24c53e1dee615067295849", + "text": "Including teams at" + }, + { + "span_id": "span-f80c3a9d62614223de020b4f", + "text": "01 Leadership" + }, + { + "span_id": "span-1ec6c1dd0c8e2b1191782ab1", + "text": "“We are at the beginning of the most significant technological shift in history." + }, + { + "span_id": "span-ef153a59b01d50b9582e47f3", + "text": "The solutions that define this era, in healthcare, energy, finance, and education, will be built by AI engineers." + }, + { + "span_id": "span-b725a0113b36c3992383f98b", + "text": "I am building Towards AI to be where those engineers are made.”" + }, + { + "span_id": "span-e79535fd6d447026477adc29", + "text": "Louie Peters · Co-founder & CEO" + }, + { + "span_id": "span-55bd756e19a0e16ba11cba02", + "text": "“I left academia to solve real problems and discovered the hardest one was finding training that reflected how production AI actually works." + }, + { + "span_id": "span-7b18a3dc00596699db2edfaf", + "text": "So I built them." + }, + { + "span_id": "span-499c7845890115da2ffc7c79", + "text": "Everything we create at Towards AI is focused on one thing: you will learn to build and ship production-grade AI systems.”" + }, + { + "span_id": "span-ab7d08f4ca3e31e9f6836e1b", + "text": "Louis-François Bouchard · Co-founder & CTO" + }, + { + "span_id": "span-d52c2563ecaf749bfa299157", + "text": "02 What we do" + }, + { + "span_id": "span-68de39910c95e54c4c6e1640", + "text": "1 / 3 For individuals" + }, + { + "span_id": "span-dca5e3742cf9205a0ea2bdd8", + "text": "The most in-depth path from developer to AI engineer Most developers hit the same wall: tutorials that work in notebooks but fall apart in production." + }, + { + "span_id": "span-e2d67ac48974c0e8bd1e461b", + "text": "Our courses are built around one project taken end-to-end, from raw data pipeline to live deployment, including the failure modes notebooks never show you." + }, + { + "span_id": "span-682a3cf40ad899e627fed0bc", + "text": "By the time you finish, you've shipped something that works." + }, + { + "span_id": "span-d32044b240fe86ecad144732", + "text": "Become an AI Engineer" + }, + { + "span_id": "span-152eb2b987bd236be79f4a3c", + "text": "2 / 3 For teams" + } + ] + }, + { + "chunk_id": "tai-c830b75484af9f65953bea0d", + "text": "In-person bootcamps\nIntensive training to accelerate AI adoption within your teams Your developers know how to code. The gap is knowing how to build reliably on top of probabilistic systems, so LLM features hold up under real usage. Our bootcamps close that gap directly, tailored to your stack and the specific problems your team is trying to solve. Get in touch 3 / 3 For organizations\n\nCustomized enterprise programs\nEnd-to-end AI consulting for organizations building with LLMs Most enterprise teams lack the production experience to know which AI initiatives are worth building. Our team of 15+ AI practitioners works directly with yours to identify the right use cases and build systems that actually deliver. Learn more Engagement formats Advisory Strategic guidance on use cases, models, and architecture Sprints Hands-on development sprints with your team Custom build LLM systems built end-to-end, handed over documented Enablement Training that keeps the capability in-house 03 Trusted voices\n\nIndustry leaders on our resources\nSince 2019\n\nAbout Towards AI\nSince 2019, we have been bridging the gap between AI research and what it actually takes to build with it in industry. What started as a newsletter has grown into one of the largest AI learning communities: 500,000+ learners, 200,000+ newsletter subscribers, and a 100,000-member Discord where engineers share what they are actually building. Behind every course, bootcamp, and enterprise program is a team of 15+ AI practitioners who work on production systems daily. They built the training they couldn't find anywhere else. 04 Who we serve", + "heading": "Customized enterprise programs", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-ea49d124cb97ecbb7b351112", + "text": "Intensive training to accelerate AI adoption within your teams Your developers know how to code." + }, + { + "span_id": "span-9e3a0c09c5e4c86334206857", + "text": "The gap is knowing how to build reliably on top of probabilistic systems, so LLM features hold up under real usage." + }, + { + "span_id": "span-1a2ee7ae6e3049c9ce799030", + "text": "Our bootcamps close that gap directly, tailored to your stack and the specific problems your team is trying to solve." + }, + { + "span_id": "span-c611af836e3cae88759f349b", + "text": "Get in touch" + }, + { + "span_id": "span-d37111b38d8dc4bf80e7b035", + "text": "3 / 3 For organizations" + }, + { + "span_id": "span-3dfcd03e4ffb19f791527941", + "text": "End-to-end AI consulting for organizations building with LLMs Most enterprise teams lack the production experience to know which AI initiatives are worth building." + }, + { + "span_id": "span-5c802e2e2d6da7be20dd265f", + "text": "Our team of 15+ AI practitioners works directly with yours to identify the right use cases and build systems that actually deliver." + }, + { + "span_id": "span-f7c0ac83f197e7f201bfb649", + "text": "Learn more" + }, + { + "span_id": "span-ce5c095b2415afbca01ca583", + "text": "Engagement formats" + }, + { + "span_id": "span-1446d2329da4f8611a513009", + "text": "Strategic guidance on use cases, models, and architecture" + }, + { + "span_id": "span-9002a934e6c8f5c73ad75345", + "text": "Hands-on development sprints with your team" + }, + { + "span_id": "span-2e3ad313a46d891ef098c4a7", + "text": "Custom build" + }, + { + "span_id": "span-1b01c0843bdc478f94168c19", + "text": "LLM systems built end-to-end, handed over documented" + }, + { + "span_id": "span-5613aa8f4cdbce4fe7dbe66a", + "text": "Training that keeps the capability in-house" + }, + { + "span_id": "span-d6572f6a808bbdf43f9b22d4", + "text": "03 Trusted voices" + }, + { + "span_id": "span-45682aa72352b37baa7968ff", + "text": "Since 2019" + }, + { + "span_id": "span-0a48063bbdd1a097f1d43a31", + "text": "Since 2019, we have been bridging the gap between AI research and what it actually takes to build with it in industry." + }, + { + "span_id": "span-5032a745f7e3629949c787a9", + "text": "What started as a newsletter has grown into one of the largest AI learning communities: 500,000+ learners, 200,000+ newsletter subscribers, and a 100,000-member Discord where engineers share what they are actually building." + }, + { + "span_id": "span-7f37ef5b7771628cd90ddae4", + "text": "Behind every course, bootcamp, and enterprise program is a team of 15+ AI practitioners who work on production systems daily." + }, + { + "span_id": "span-a8647451a459507fdd7c500b", + "text": "They built the training they couldn't find anywhere else." + }, + { + "span_id": "span-b38d5bc4e3903a2257633972", + "text": "04 Who we serve" + } + ] + }, + { + "chunk_id": "tai-afd2c4500657f692cb909e2f", + "text": "Who are these resources for?\nOur material is designed for highly motivated learners who are ready to invest in their future. We offer courses for every skill level, and none of them are passive: you'll be actively building through hands-on projects rather than sitting through long videos. Whether you're a beginner or an experienced coder, the content challenges you and grows with your skills. Expect to dedicate real time to practical assignments that reflect production applications. Our learners have moved into high-paying AI roles and led AI initiatives inside their companies by gaining the skills to build and operate advanced AI systems. Start here\n\nFind your course\nFive programs, one pathway: from beginner Python to full stack AI engineering and production agents. Browse the pathway Talk to us about team training", + "heading": "Who are these resources for?", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-0bb88e7a6ab370a8c4e197ae", + "text": "Our material is designed for highly motivated learners who are ready to invest in their future." + }, + { + "span_id": "span-d78fc7d404ac7f47a7fa1bd7", + "text": "We offer courses for every skill level, and none of them are passive: you'll be actively building through hands-on projects rather than sitting through long videos." + }, + { + "span_id": "span-48d35f12685176d0d183cb39", + "text": "Whether you're a beginner or an experienced coder, the content challenges you and grows with your skills." + }, + { + "span_id": "span-da4419f7b45c88eb236d5386", + "text": "Expect to dedicate real time to practical assignments that reflect production applications." + }, + { + "span_id": "span-829cd2054c345e24417985fa", + "text": "Our learners have moved into high-paying AI roles and led AI initiatives inside their companies by gaining the skills to build and operate advanced AI systems." + }, + { + "span_id": "span-d5d99e90f9ebf105e69d157d", + "text": "Start here" + }, + { + "span_id": "span-3fae1e1365ce20e56efa4471", + "text": "Five programs, one pathway: from beginner Python to full stack AI engineering and production agents." + }, + { + "span_id": "span-1363a124bb937ec3246059d2", + "text": "Browse the pathway Talk to us about team training" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "0ea004e46b9b2ec3401c6a9f339234196c09f4e598ce491dc3a2450082e0147c", + "content_hash": "0ea004e46b9b2ec3401c6a9f339234196c09f4e598ce491dc3a2450082e0147c", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/about/" + ], + "evidence_hash": "cf19c7179ed37752111c5f039c8910a0a07b07827ddba6e11371c60c5e325b51" + }, + { + "discovered_url": "https://towardsai.com/academy/contact/", + "url": "https://towardsai.com/academy/contact/", + "canonical_url": "https://towardsai.com/academy/contact/", + "host": "towardsai.com", + "path": "/academy/contact", + "kind": "page", + "offer_id": null, + "entity_id": "page:towardsai.com/academy/contact", + "title": "Contact Us · Towards AI Academy", + "review_title": "Contact", + "reviewed_summary": "Towards AI Academy contact options.", + "meta_description": "Talk to the Towards AI team about courses, membership, team training, partnerships, or support. We reply within one business day.", + "headings": [ + "Talk to the Towards AI team", + "Pick the right door", + "30-day money-back guarantee on all courses" + ], + "text": "Contact us\n\nTalk to the Towards AI team\nQuestions about courses, membership, team training, partnerships, or your account. Every message is read by the team, and we reply within one business day. 01 Fastest routes\n\nPick the right door\n01 Course or membership questions. Use the form and pick the closest topic. Discounts, bundles, and refunds included. 02 Already a student? Your courses, certificates, and billing live in your student dashboard . Course-specific help is fastest in your course's Discord channel. 03 Team training and enterprise. See the AI engineer programme , the agentic developer programme , or enterprise enablement . 04 Affiliates and referrals. Tiers, payouts, and signup are on the affiliate page . 05 Community. Join 100k+ engineers in our Discord for discussion, help, and study groups. Full name Email Topic Course question Membership Refund or billing Team training / B2B Affiliate or partnership Press Something else Your message I agree to be contacted about this request. Send message Refunds\n\n30-day money-back guarantee on all courses\nIf a course isn't right for you, tell us within 30 days for a full refund. No hassle, no worries. Request via the form", + "links": [ + { + "text": "your student dashboard", + "url": "https://academy.towardsai.net/enrollments" + }, + { + "text": "AI engineer programme", + "url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/" + }, + { + "text": "agentic developer programme", + "url": "https://towardsai.com/enterprise/agentic-developer-conversion/" + }, + { + "text": "enterprise enablement", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + }, + { + "text": "Discord", + "url": "https://discord.gg/learnaitogether" + } + ], + "chunks": [ + { + "chunk_id": "tai-c8e1008f64af4eb02fb5e174", + "text": "Contact us\n\nTalk to the Towards AI team\nQuestions about courses, membership, team training, partnerships, or your account. Every message is read by the team, and we reply within one business day. 01 Fastest routes\n\nPick the right door\n01 Course or membership questions. Use the form and pick the closest topic. Discounts, bundles, and refunds included. 02 Already a student? Your courses, certificates, and billing live in your student dashboard . Course-specific help is fastest in your course's Discord channel. 03 Team training and enterprise. See the AI engineer programme , the agentic developer programme , or enterprise enablement . 04 Affiliates and referrals. Tiers, payouts, and signup are on the affiliate page . 05 Community. Join 100k+ engineers in our Discord for discussion, help, and study groups. Full name Email Topic Course question Membership Refund or billing Team training / B2B Affiliate or partnership Press Something else Your message I agree to be contacted about this request. Send message Refunds\n\n30-day money-back guarantee on all courses\nIf a course isn't right for you, tell us within 30 days for a full refund. No hassle, no worries. Request via the form", + "heading": "Pick the right door", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-727416ba0930ec6286f023f2", + "text": "Contact us" + }, + { + "span_id": "span-f378f7d04f0ff919de8bd45c", + "text": "Questions about courses, membership, team training, partnerships, or your account." + }, + { + "span_id": "span-76354241dde80637ecf46542", + "text": "Every message is read by the team, and we reply within one business day." + }, + { + "span_id": "span-fa64ba1a77c098b194341dbc", + "text": "01 Fastest routes" + }, + { + "span_id": "span-232d80bab481edd43ecec8b5", + "text": "01 Course or membership questions." + }, + { + "span_id": "span-0f302b9eb4e40f8ebe710de1", + "text": "Use the form and pick the closest topic." + }, + { + "span_id": "span-3e06747f0c55054f0e7d013b", + "text": "Discounts, bundles, and refunds included." + }, + { + "span_id": "span-84b5750cedfd296e7bdfd5cc", + "text": "02 Already a student?" + }, + { + "span_id": "span-40ee4d1256d96c4e71219ca1", + "text": "Your courses, certificates, and billing live in your student dashboard ." + }, + { + "span_id": "span-7bfaaabc5d67e865905b977b", + "text": "Course-specific help is fastest in your course's Discord channel." + }, + { + "span_id": "span-020a54253a25bb82dbfd5cb8", + "text": "03 Team training and enterprise." + }, + { + "span_id": "span-218dc7d673d77da9d1507c32", + "text": "See the AI engineer programme , the agentic developer programme , or enterprise enablement ." + }, + { + "span_id": "span-6e1bd5b3b2574de42747e047", + "text": "04 Affiliates and referrals." + }, + { + "span_id": "span-35ab01b85d3c474f73b438c6", + "text": "Tiers, payouts, and signup are on the affiliate page ." + }, + { + "span_id": "span-19f1837638f9bbc937a2de49", + "text": "05 Community." + }, + { + "span_id": "span-6fc6781a7b8ed5332797c094", + "text": "Join 100k+ engineers in our Discord for discussion, help, and study groups." + }, + { + "span_id": "span-5a87f28ce9de98b8ecfa8c26", + "text": "Full name Email Topic Course question Membership Refund or billing Team training / B2B Affiliate or partnership Press Something else Your message I agree to be contacted about this request." + }, + { + "span_id": "span-6a4d6bcf33c60e0d514e95c3", + "text": "Send message" + }, + { + "span_id": "span-1ef1b9889dd17eedee7d6124", + "text": "30-day money-back guarantee on all courses" + }, + { + "span_id": "span-7e8734b1d651d24c00d6e2fd", + "text": "If a course isn't right for you, tell us within 30 days for a full refund." + }, + { + "span_id": "span-5562c0fb869fb63fa1f21d49", + "text": "No hassle, no worries." + }, + { + "span_id": "span-c9b972f02695de0b822ff4a1", + "text": "Request via the form" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "812e474602b87e136514ab1a17328a791a6a9b92b1255c8f6340322bf6287cb6", + "content_hash": "812e474602b87e136514ab1a17328a791a6a9b92b1255c8f6340322bf6287cb6", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/contact/" + ], + "evidence_hash": "a68ff4bfbde8acac2999c39ecebc5a504f5b58a887ed897879797539be1e73af" + }, + { + "discovered_url": "https://towardsai.com/contribute/", + "url": "https://towardsai.com/contribute/", + "canonical_url": "https://towardsai.com/contribute/", + "host": "towardsai.com", + "path": "/contribute", + "kind": "community", + "offer_id": null, + "entity_id": "page:towardsai.com/contribute", + "title": "Write for Towards AI · Contribute to the Leading AI Publication", + "review_title": "", + "reviewed_summary": "", + "meta_description": "Publish your technical writing with Towards AI and reach 500k+ AI practitioners. Submit a draft; we reply within 72 hours. Accepted writers join our network and submit directly.", + "headings": [ + "Publish with the leading AI publication", + "From draft to published", + "Read the guidelines, then submit", + "We review within 72 hours", + "Accepted writers submit directly", + "Your name on work that gets read", + "500k+ AI practitioners", + "A real editorial process", + "A public body of technical work", + "Send us your draft", + "Not ready to publish yet?" + ], + "text": "Write for Towards AI\n\nPublish with the leading AI publication\nPut your technical writing in front of 500k+ AI practitioners. Submit a draft below; we reply within 72 hours, and accepted writers join our network with direct submission access. Submit a draft Read the guidelines 01 How it works\n\nFrom draft to published\n01\n\nRead the guidelines, then submit\nMake sure your draft meets our guidelines in full, then send it through the form below with a link to the draft. 02\n\nWe review within 72 hours\nEvery submission is read by our editorial team. We receive several per day, so here's the deal: if you haven't heard from us within 72 hours, consider the piece not accepted this time. Please don't let that discourage future drafts. 03\n\nAccepted writers submit directly\nOnce accepted, you join the Towards AI writers network and can submit articles straight to our queue for future pieces. 02 Why write with us\n\nYour name on work that gets read\nReach\n\n500k+ AI practitioners\nThe Towards AI publication is one of the largest AI publications on Medium, backed by a newsletter and community that put technical writing in front of the people building with AI. Editing\n\nA real editorial process\nAccepted pieces go through our content team's review and editing workflow, so what ships under your name is sharper than what you submitted. Career proof\n\nA public body of technical work\nA few published articles build a discoverable portfolio of technical writing with your name on it, the kind employers and collaborators actually find. 03 Submit\n\nSend us your draft\nShare a link to the draft (Google Docs, Medium draft link, or similar) along with a little about you. We reply within 24 to 72 hours if the piece is a fit. By submitting you confirm you've read the submission guidelines . Full name Email Link to your draft article Twitter / X profile (optional) Link to your CV (optional) Message I agree to be contacted about this submission. Submit draft More ways in\n\nNot ready to publish yet?\nRead what our writers ship, join the community, or sharpen your skills first. Read the publication Join the Discord Browse courses", + "links": [ + { + "text": "Read the guidelines", + "url": "https://pub.towardsai.net/submit-your-medium-story-to-towards-ai-a4fa7e8b141d" + }, + { + "text": "our guidelines", + "url": "https://pub.towardsai.net/submit-your-medium-story-to-towards-ai-a4fa7e8b141d" + }, + { + "text": "submit articles straight to our queue", + "url": "https://ws.towardsai.net/submit" + }, + { + "text": "submission guidelines", + "url": "https://pub.towardsai.net/submit-your-medium-story-to-towards-ai-a4fa7e8b141d" + }, + { + "text": "Read the publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Join the Discord", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Browse courses", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-9644108d4792ed847049113e", + "text": "Write for Towards AI\n\nPublish with the leading AI publication\nPut your technical writing in front of 500k+ AI practitioners. Submit a draft below; we reply within 72 hours, and accepted writers join our network with direct submission access. Submit a draft Read the guidelines 01 How it works\n\nFrom draft to published\n01\n\nRead the guidelines, then submit\nMake sure your draft meets our guidelines in full, then send it through the form below with a link to the draft. 02\n\nWe review within 72 hours\nEvery submission is read by our editorial team. We receive several per day, so here's the deal: if you haven't heard from us within 72 hours, consider the piece not accepted this time. Please don't let that discourage future drafts. 03\n\nAccepted writers submit directly\nOnce accepted, you join the Towards AI writers network and can submit articles straight to our queue for future pieces. 02 Why write with us\n\nYour name on work that gets read\nReach\n\n500k+ AI practitioners\nThe Towards AI publication is one of the largest AI publications on Medium, backed by a newsletter and community that put technical writing in front of the people building with AI. Editing\n\nA real editorial process\nAccepted pieces go through our content team's review and editing workflow, so what ships under your name is sharper than what you submitted. Career proof\n\nA public body of technical work\nA few published articles build a discoverable portfolio of technical writing with your name on it, the kind employers and collaborators actually find. 03 Submit", + "heading": "Publish with the leading AI publication", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-4107e5f0324571c7e6365745", + "text": "Write for Towards AI" + }, + { + "span_id": "span-b58f1168aa5045f15e2e8414", + "text": "Put your technical writing in front of 500k+ AI practitioners." + }, + { + "span_id": "span-c51ca7f70071a3fef16e8afd", + "text": "Submit a draft below; we reply within 72 hours, and accepted writers join our network with direct submission access." + }, + { + "span_id": "span-9dece507dfe89160460a1a5c", + "text": "Submit a draft Read the guidelines" + }, + { + "span_id": "span-de6b9d925a5e2a6b5bf6a014", + "text": "01 How it works" + }, + { + "span_id": "span-1bcfdf3aa2d216e78fd80a6f", + "text": "From draft to published" + }, + { + "span_id": "span-9de9761146a18dca6822f9e8", + "text": "Make sure your draft meets our guidelines in full, then send it through the form below with a link to the draft." + }, + { + "span_id": "span-20f1e4a2a34c7fc7975b3163", + "text": "We review within 72 hours" + }, + { + "span_id": "span-fca817c835d6c07be6aa737f", + "text": "Every submission is read by our editorial team." + }, + { + "span_id": "span-d7104cca0fcabd019ec4b1f9", + "text": "We receive several per day, so here's the deal: if you haven't heard from us within 72 hours, consider the piece not accepted this time." + }, + { + "span_id": "span-2f1041097b402c6951e2effc", + "text": "Please don't let that discourage future drafts." + }, + { + "span_id": "span-e428f67b9031e0c8505e0d5d", + "text": "Once accepted, you join the Towards AI writers network and can submit articles straight to our queue for future pieces." + }, + { + "span_id": "span-37ca804f9984f76750e1c69d", + "text": "02 Why write with us" + }, + { + "span_id": "span-e88960d0adca418a62e1624f", + "text": "Your name on work that gets read" + }, + { + "span_id": "span-10fbb6b0d13daad53b781457", + "text": "500k+ AI practitioners" + }, + { + "span_id": "span-917084534d3568bcc62fb45c", + "text": "The Towards AI publication is one of the largest AI publications on Medium, backed by a newsletter and community that put technical writing in front of the people building with AI." + }, + { + "span_id": "span-7ae147a88782d90b2b42b592", + "text": "Accepted pieces go through our content team's review and editing workflow, so what ships under your name is sharper than what you submitted." + }, + { + "span_id": "span-fb30cf3b69f64cf0532c863b", + "text": "Career proof" + }, + { + "span_id": "span-b225c1d79baf287bb2d37c07", + "text": "A few published articles build a discoverable portfolio of technical writing with your name on it, the kind employers and collaborators actually find." + }, + { + "span_id": "span-c218bc7279d841d834de7ae8", + "text": "03 Submit" + } + ] + }, + { + "chunk_id": "tai-d097fb7a2ac5bcbe95cdf477", + "text": "Send us your draft\nShare a link to the draft (Google Docs, Medium draft link, or similar) along with a little about you. We reply within 24 to 72 hours if the piece is a fit. By submitting you confirm you've read the submission guidelines . Full name Email Link to your draft article Twitter / X profile (optional) Link to your CV (optional) Message I agree to be contacted about this submission. Submit draft More ways in\n\nNot ready to publish yet?\nRead what our writers ship, join the community, or sharpen your skills first. Read the publication Join the Discord Browse courses", + "heading": "Send us your draft", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-47dfac3a529ced0a97029ce8", + "text": "Share a link to the draft (Google Docs, Medium draft link, or similar) along with a little about you." + }, + { + "span_id": "span-0251f0b14161045e4cbfc772", + "text": "We reply within 24 to 72 hours if the piece is a fit." + }, + { + "span_id": "span-0f6469c20e9f2e130e2f083c", + "text": "By submitting you confirm you've read the submission guidelines ." + }, + { + "span_id": "span-924ff3b7f85fd74f6468b7d0", + "text": "Full name Email Link to your draft article Twitter / X profile (optional) Link to your CV (optional) Message I agree to be contacted about this submission." + }, + { + "span_id": "span-7bc7bd5c4e80eb39bd43223a", + "text": "Submit draft" + }, + { + "span_id": "span-c359e0e54321f51d626de360", + "text": "More ways in" + }, + { + "span_id": "span-15463cce71fd12fd35a61813", + "text": "Read what our writers ship, join the community, or sharpen your skills first." + }, + { + "span_id": "span-56ac107dd92d560bac69da35", + "text": "Read the publication Join the Discord Browse courses" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "671a4e7df6e18a0979991d3540cf23c00bd589819caa03de8f11b09c26dc9554", + "content_hash": "671a4e7df6e18a0979991d3540cf23c00bd589819caa03de8f11b09c26dc9554", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/contribute/" + ], + "evidence_hash": "459280a5f3ba797a5a7c7e687dd3df06cce3853eec5f289c8f608248371c590a" + }, + { + "discovered_url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/", + "url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/", + "canonical_url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/", + "host": "towardsai.com", + "path": "/enterprise/software-developer-to-ai-engineer", + "kind": "b2b", + "offer_id": null, + "entity_id": "page:towardsai.com/enterprise/software-developer-to-ai-engineer", + "title": "AI Engineer Bootcamp for Enterprises: Convert Your Developers | Towards AI", + "review_title": "Software Developer to AI Engineer", + "reviewed_summary": "Enterprise AI-engineer conversion program.", + "meta_description": "You cannot hire AI engineers at the pace you need; convert the developers you already employ. Cohort waves, a senior residency, and per-seat licences, certified against shipped capstones. Book a scoping call.", + "headings": [ + "You can't hire AI engineers at the pace you need. Convert them.", + "Three ways to build the capability in-house", + "AI engineer conversion cohort", + "AI leader residency", + "Enterprise Academy licence", + "Two steps: use AI to build software, then build AI into it", + "Deployed AI pays back. Pilots without owners don't.", + "The questions your board will ask", + "The curriculum , module by module", + "Every pattern we teach appears in a system we built", + "We ran this conversion line on ourselves first", + "Evidence before scale", + "Scoping call · 30–60 min", + "Baseline & co-design", + "Pilot cohort", + "Certify & scale decision", + "The winners of the next three years convert their existing developer base fastest.", + "Frequently asked questions", + "Book a scoping call" + ], + "text": "Step 2 of 2 · Selected builders\n\nYou can't hire AI engineers at the pace you need. Convert them.\nYour developers already hold the domain and codebase knowledge your AI systems will run on. We convert the strongest into AI engineers who ship production LLM systems. Book a scoping call See the formats 30 minutes with the instructors who deliver it. No sales team. +143% growth in AI engineer postings, the #1 fastest-growing US job (LinkedIn 2026) 56% wage premium for AI skills, up from 25% (PwC), and the labs hire each other first 95% of deployed AI meets or exceeds ROI expectations (McKinsey); deployment is the gap Programme at a glance Core Conversion cohorts , part-time around delivery Leads 8–12-week senior residency, 1 to 6 per wave Scale Per-seat Academy licence, volume bands Taught by 15+ practising AI engineers Certified Against shipped capstones, not attendance Before this Most teams run Step 1 first; it selects the cohort Training delivered for teams at 01 The formats\n\nThree ways to build the capability in-house\nEvery format ends in certification against shipped work and a written capability debrief. We recommend one against your baseline in the scoping call. Cohort waves Core programme\n\nAI engineer conversion cohort\nSelected builders, typically your top 10–20%, progress part-time around delivery: self-paced study, live instructor checkpoints, exercises against your approved use cases, and a capstone that ships: a deployed, evaluated LLM system. Scale Waves of 10 to 40; parallel streams at scale Covers Model behaviour, RAG, evaluation, agents, deployment Leaves with Shipped capstones, certification, capability debrief Per cohort · quoted after scoping Scope a cohort → 8–12 weeks For your leads\n\nAI leader residency\nOne to six future AI leaders work with our senior architects: weekly coaching, applied labs, architecture reviews, and a capstone from a real internal workflow. Each resident turns one workflow into a reference build later cohorts train on. Scale 1 to 6 senior engineers per wave Covers Workflow qualification, baselines, adoption, the demo-to-production gap Leaves with Reference builds, internal instructors, your standards Per resident · priced as senior design work Nominate residents → Per seat\n\nEnterprise Academy licence\nThe full Academy bundle: Full Stack AI Engineering (92 lessons), Agentic AI Engineering (5.0-rated), the Agentic Development course, and AI for Work, with live kickoff, monthly mentorship clinics, and leadership analytics. Lifetime updates as our client work evolves. Scale Hundreds to thousands of seats Covers The complete curriculum, self-paced with role paths Leaves with One shared foundation and progress analytics Per seat · volume bands Discuss seats → At 1,000+ developers: residency seed → 20–40-seat pilot → waves of 50–250 in parallel streams �� top graduates co-instruct as internal champions. Per-developer cost drops as the programme scales. A Python foundations route exists for mixed-skill cohorts. 02 The system\n\nTwo steps: use AI to build software, then build AI into it\nStep 1 makes every developer effective with coding agents, and its debrief names who belongs here. You are on the Step 2 page. Step 1 · Agentic developers Step 2 · AI engineers ← this page Who Every software developer, no AI background needed Selected builders , typically the top 10–20% They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems Outcome Ship faster on today's roadmap The capability you cannot hire : AI products and systems, built in-house Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · 8–12-week residency · per-seat licence When Now: it pays back on work already in flight After Step 1, or now if your devs already work seriously with agents Not there yet? Start with Step 1 → Same instructors, same standard, one scoping call covers both. 03 Why convert\n\nDeployed AI pays back. Pilots without owners don't.\nThe pilots graveyard MIT's 2025 study attributes the failure of ~95% of enterprise GenAI pilots to a “learning gap”: workflow and people integration, not model quality. S&P Global finds 42% of enterprises abandoned most AI initiatives in 2025, up from 17%. Pilots stall because nobody owns production: design, evaluation, deployment, operations. MIT NANDA 2025 · S&P Global 2025 Deployment pays McKinsey finds only 7% of firms have deployed AI organisation-wide while 65% are stuck at proof of concept, yet 95% of deployed AI meets or exceeds ROI expectations, and 94% of firms are increasing spend into exactly that gap. The missing input isn't budget or models. It's people who can carry systems to production. McKinsey 2026 You can't hire your way out AI engineer is the #1 fastest-growing US job, postings up 143% year on year, with a 56% wage premium for AI skills, and LinkedIn's own data shows the most common feeder role is the software engineer. Your conversion pipeline is already on payroll; a new hire spends a year learning what your team already knows. LinkedIn 2026 · PwC 2025 Capability compounds The first trained cohort sets the engineering standard every later cohort inherits. Graduates seed later waves as internal champions and co-instructors, per-developer cost drops as the programme scales, and the capability stays in-house, with no dependence on us for the next use case. Programme design · scale path above 04 Straight answers\n\nThe questions your board will ask\n“Why not just hire AI engineers?” The role is barely three years old, supply is thin, the labs hire each other first, and frontier comp is extreme. Meanwhile your developers already hold the domain, systems, and codebase knowledge AI systems run on. Conversion wins on cost, speed, retention, and context. “Most AI initiatives fail anyway.” They fail for skills reasons: MIT attributes the 95% pilot-failure rate to a learning gap, not model quality, while 95% of deployed AI meets or exceeds ROI expectations (McKinsey). Capability-building is precisely the variable that moves you across that divide. “Training doesn't stick.” Lecture-style training doesn't. This programme certifies against shipped capstones (a deployed, evaluated LLM system per engineer) with exercises drawn from your approved roadmap, and closes with a written capability debrief. Nobody is certified for attending. “We can't take engineers off delivery.” Conversion runs part-time around delivery : self-paced study plus scheduled live checkpoints, shaped with engineering leadership. The residency is a defined weekly commitment for one to six seniors, not weeks out of the roadmap. “How do we choose who to convert?” Run Step 1 first: its capability debrief names the builders ready for this programme. Otherwise we baseline and segment the population in scoping (developers, data scientists, existing agent users) and agree the first cohort with engineering leadership. “Generic courses won't fit our stack.” Depth and emphasis are agreed against your engineers' baseline; electives are added where your stack needs them; exercises come from your roadmap, never generic demos. Every pattern we teach appears in a system we built ; see below. 05 Full transparency\n\nThe curriculum , module by module\nDepth and emphasis are agreed against your engineers' baseline in scoping. Exercises come from your approved roadmap, never generic demos. 01 Foundations and model behaviour core + 1 Models, prompting, and the limitations that shape engineering decisions 2 Security, prompt injection, and responsible technical use 3 Choosing the right technique for the task: prompting, RAG, fine-tuning, or agents 02 Systems: retrieval and application patterns build + 1 Retrieval-augmented generation, from first pipeline to production patterns 2 Framework and architecture trade-offs; context and application design 3 Application-oriented exercises against approved use cases 03 Evaluation and reliability rigour + 1 Test-set design and evaluation as an engineering discipline 2 Generation and retrieval evaluation; LLM-as-judge and regression thinking 3 Observability, so output is never accepted on intuition alone 04 Agent engineering advanced + 1 Agentic patterns and when agents beat simpler techniques; tool use and function calling 2 Orchestration and multi-agent systems; MCP integrations; memory and state design 3 Agent evaluation, guardrails, and human-review boundaries 05 Optimisation and deployment production + 1 Fine-tuning and optimisation; deployment and cost control 2 Monitoring in production, and staying current as the field moves 3 Electives where your stack needs them: large codebases, domain integrations, compliance 06 Scoped to your baseline custom + 1 Depth and emphasis agreed against the engineers' baseline during scoping 2 Module weighting follows your use cases, not a fixed timetable 3 No stage is filler for the next 07 Capstone: certified against shipped work certified + 1 Every module lands in capstone work: a deployed, evaluated LLM system 2 Certification against shipped capstones, not attendance 3 A written capability debrief closes each cohort Outcome: engineers who design, evaluate, ship, and maintain production LLM systems, certified against a deployed capstone. 06 We teach what we ship\n\nEvery pattern we teach appears in a system we built\n21+ AI applications shipped to production across finance, government, law enforcement, healthcare, and industry. The curriculum is drawn from that work, which is why your training partner must be a practising builder. Banker-reasoning engine Banking · Europe Agentic financial-intelligence platform routing 1M+ words per company across four frontier model families; ~800 prompts and 20-step workflows per analysis. Teaches: routing, orchestration, and evaluation, from idea to live demos with global banks. Dialog Insight Martech · Canada Multi-month production-agent enablement: weekly advisory, 16+ hours of guided code transfer, six runnable reference repositories, a 25-scenario benchmark. Outcome: working code the team owns: a production-shaped agent stack, runnable and auditable. Document extraction at scale Mortgage · Canada 50+ bilingual document types for underwriting; three independent extraction passes with consensus; never-overwrite Salesforce integration; data-residency compliant. Teaches: structured outputs, consensus architectures, human-review boundaries. “The most comprehensive textbook to date on building LLM applications; every essential topic in an AI engineer's toolkit.” Jerry Liu Founder & CEO, LlamaIndex, on our book “An indispensable guide for anyone venturing into the world of LLMs, covering everything from theory to practical deployment.” Senior Data Engineer Meta “Invaluable to anyone looking to dive into the field quickly and efficiently.” Senior Applied Research Scientist Mila (Yoshua Bengio's AI lab) 07 Proven on ourselves\n\nWe ran this conversion line on ourselves first\nOur delivery arm needed more production AI engineers than the market could supply, so we built a conversion line: Learner → Author → Instructor → Engineer. 15+ of our own engineers came off it. Work is the audition, and your cohorts are held to the same standard. LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production (10k+ copies). Chief AI Officer for a banking agentic-engine build. Ex-J.P. Morgan VP, Research. LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA (Yoshua Bengio's lab). Built our 100k+ member community. FC Fabio Chiusano Senior AI Engineer NLP lead; ex-founder & Head of Data Science. Joined as a contributing author; proof the pipeline works. AK Assel Kashkenbayeva Deployment lead Ex-Palantir project lead; UCL PhD. Carries the forward-deployment model the residency teaches. 500k+ AI practitioners taught since 2019 15+ in-house AI engineers, grown from our own line 21+ AI applications shipped to production 10k+ copies of Building LLMs for Production sold 08 How it runs\n\nEvidence before scale\n01\n\nScoping call · 30–60 min\nTeam sizes, stacks, baseline, and target capabilities, discussed with the instructors. We'll tell you if Step 1 should come first. 02\n\nBaseline & co-design\nWe assess and segment the developer population; a nominee per team keeps materials resonant with real work. Agreed with engineering leadership before anything is booked. 03\n\nPilot cohort\nTypically 20 to 40 engineers, chosen with leadership. Exercises from your approved roadmap, never generic demos. Capstones build throughout. 04\n\nCertify & scale decision\nCertification against shipped capstones and a written debrief of what the cohort can now do. We scale to further waves only when the debrief earns it. + Pilot first. One cohort, agreed success measures, and a scale-decision date, before any programme commitment. + You own the artefacts. Capstones, standards, and reference builds stay in-house. No dependence on us for the next use case. + Procurement-ready. NDAs standard, PO and invoicing supported, no production access required; your code stays yours. + Portfolio roll-outs. For groups and PE portfolios: shared standards, per-company scoping, champions carried across companies.\n\nThe winners of the next three years convert their existing developer base fastest.\nOne call, one plan, one pilot. Scale only when the debrief earns it. Book a scoping call Start with Step 1 09 Logistics\n\nFrequently asked questions\nDo participants need to do Step 1 first? + It's the recommended order, not a hard rule. Step 1 builds the agent fluency this programme assumes, and its debrief tells you who to send here. If your developers already work seriously with coding agents, we confirm the baseline in scoping and start directly. Hard prerequisites: working Python and software engineering experience; a Python foundations route exists for mixed cohorts. Where do capstones fit if delivery is part-time? + Cohorts run over weeks around delivery: self-paced study, live checkpoints, and capstone work that builds throughout. Certification happens when the capstone ships (a deployed, evaluated system), not when the calendar runs out. Does \"conversion\" mean a guaranteed role change? + No. The programme develops applied AI engineering capability in existing software developers, certified against shipped capstone work. Scope, assessment, and outcomes are agreed per engagement. What proof can we review before buying? + The curriculum is on this page, the underlying courses have public previews, our book is in print, and the production systems behind the curriculum are described above. References and the most relevant engagements are discussed in scoping. 10 Next step\n\nBook a scoping call\nTell us about your engineering team, current baseline, and the AI capabilities you want to build. We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the programme. 1 On the call: team sizes, stacks, baseline, target capabilities, and which formats fit which teams. 30–60 minutes, with instructors. 2 You leave with: a recommended programme architecture, workload estimate, commercial range, and measurement plan. 3 No pressure: if Step 1, or nothing, is the right answer, we say so. Every enquiry is read by the Towards AI team. Prefer email? louis@towardsai.net Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet AI engineer conversion cohort AI leader residency Enterprise Academy licence Step 1: Agentic developer programme What does your team need? I agree to be contacted about this request. Book a scoping call", + "links": [ + { + "text": "Not there yet? Start with Step 1 →", + "url": "https://towardsai.com/enterprise/agentic-developer-conversion/" + }, + { + "text": "Step 1", + "url": "https://towardsai.com/enterprise/agentic-developer-conversion/" + }, + { + "text": "Start with Step 1", + "url": "https://towardsai.com/enterprise/agentic-developer-conversion/" + } + ], + "chunks": [ + { + "chunk_id": "tai-ef389ffd604c3c6ae9fa4a54", + "text": "Step 2 of 2 · Selected builders\n\nYou can't hire AI engineers at the pace you need. Convert them.\nYour developers already hold the domain and codebase knowledge your AI systems will run on. We convert the strongest into AI engineers who ship production LLM systems. Book a scoping call See the formats 30 minutes with the instructors who deliver it. No sales team. +143% growth in AI engineer postings, the #1 fastest-growing US job (LinkedIn 2026) 56% wage premium for AI skills, up from 25% (PwC), and the labs hire each other first 95% of deployed AI meets or exceeds ROI expectations (McKinsey); deployment is the gap Programme at a glance Core Conversion cohorts , part-time around delivery Leads 8–12-week senior residency, 1 to 6 per wave Scale Per-seat Academy licence, volume bands Taught by 15+ practising AI engineers Certified Against shipped capstones, not attendance Before this Most teams run Step 1 first; it selects the cohort Training delivered for teams at 01 The formats\n\nThree ways to build the capability in-house\nEvery format ends in certification against shipped work and a written capability debrief. We recommend one against your baseline in the scoping call. Cohort waves Core programme\n\nAI engineer conversion cohort\nSelected builders, typically your top 10–20%, progress part-time around delivery: self-paced study, live instructor checkpoints, exercises against your approved use cases, and a capstone that ships: a deployed, evaluated LLM system. Scale Waves of 10 to 40; parallel streams at scale Covers Model behaviour, RAG, evaluation, agents, deployment Leaves with Shipped capstones, certification, capability debrief Per cohort · quoted after scoping Scope a cohort → 8–12 weeks For your leads", + "heading": "You can't hire AI engineers at the pace you need. Convert them.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-b21aacec9d6ed27759c9d84f", + "text": "Step 2 of 2 · Selected builders" + }, + { + "span_id": "span-d1dc3e6fa657d4b71e70158c", + "text": "Your developers already hold the domain and codebase knowledge your AI systems will run on." + }, + { + "span_id": "span-a09b90c6e495ba9309cdfc35", + "text": "We convert the strongest into AI engineers who ship production LLM systems." + }, + { + "span_id": "span-f71dd332c0dd62c9b1d53bde", + "text": "Book a scoping call See the formats" + }, + { + "span_id": "span-05a6df733cc7f42aa3e836ec", + "text": "30 minutes with the instructors who deliver it." + }, + { + "span_id": "span-35bf531a9dd7bf2c5078d366", + "text": "No sales team. +143% growth in AI engineer postings, the #1 fastest-growing US job (LinkedIn 2026)" + }, + { + "span_id": "span-801507fba3cb4a858100177e", + "text": "56% wage premium for AI skills, up from 25% (PwC), and the labs hire each other first" + }, + { + "span_id": "span-a008d3aa4d69e596c539f401", + "text": "95% of deployed AI meets or exceeds ROI expectations (McKinsey); deployment is the gap" + }, + { + "span_id": "span-4b5dc7f169c98fdd3b97b904", + "text": "Programme at a glance" + }, + { + "span_id": "span-2058c63510b9d50ee170e9d8", + "text": "Conversion cohorts , part-time around delivery" + }, + { + "span_id": "span-8cc5d16efdc87b46b1afa041", + "text": "8–12-week senior residency, 1 to 6 per wave" + }, + { + "span_id": "span-8540a3da544d7cfefb2f2b39", + "text": "Per-seat Academy licence, volume bands" + }, + { + "span_id": "span-9827eca487a8f5dc72f66864", + "text": "Taught by" + }, + { + "span_id": "span-cc92bf0c7d45f9a2e3dfdc71", + "text": "15+ practising AI engineers" + }, + { + "span_id": "span-e58e378c3a3695213a654619", + "text": "Against shipped capstones, not attendance" + }, + { + "span_id": "span-77c3ecbb83a7501da252773e", + "text": "Before this" + }, + { + "span_id": "span-43d336f2e67f9d375ca6313b", + "text": "Most teams run Step 1 first; it selects the cohort" + }, + { + "span_id": "span-f0016c034185d49103c32f91", + "text": "Training delivered for teams at" + }, + { + "span_id": "span-1c7b3a6772a5bc47294e46bc", + "text": "01 The formats" + }, + { + "span_id": "span-9ecd4fc5457a22fe4bef81cc", + "text": "Every format ends in certification against shipped work and a written capability debrief." + }, + { + "span_id": "span-ac89b9fc6022822365c2e76a", + "text": "We recommend one against your baseline in the scoping call." + }, + { + "span_id": "span-f87613670a4b86c8947dfd6c", + "text": "Cohort waves Core programme" + }, + { + "span_id": "span-8915eb77677b1f0879f26050", + "text": "Selected builders, typically your top 10–20%, progress part-time around delivery: self-paced study, live instructor checkpoints, exercises against your approved use cases, and a capstone that ships: a deployed, evaluated LLM system." + }, + { + "span_id": "span-b8947dfaecda7fa7293ca0ea", + "text": "Scale Waves of 10 to 40; parallel streams at scale Covers Model behaviour, RAG, evaluation, agents, deployment Leaves with Shipped capstones, certification, capability debrief" + }, + { + "span_id": "span-edfd91f4c2c8f26ee253987f", + "text": "Per cohort · quoted after scoping Scope a cohort →" + }, + { + "span_id": "span-20a757a1d738ed48f4712471", + "text": "8–12 weeks For your leads" + } + ] + }, + { + "chunk_id": "tai-0b46d95c2f076942ec105587", + "text": "AI leader residency\nOne to six future AI leaders work with our senior architects: weekly coaching, applied labs, architecture reviews, and a capstone from a real internal workflow. Each resident turns one workflow into a reference build later cohorts train on. Scale 1 to 6 senior engineers per wave Covers Workflow qualification, baselines, adoption, the demo-to-production gap Leaves with Reference builds, internal instructors, your standards Per resident · priced as senior design work Nominate residents → Per seat\n\nEnterprise Academy licence\nThe full Academy bundle: Full Stack AI Engineering (92 lessons), Agentic AI Engineering (5.0-rated), the Agentic Development course, and AI for Work, with live kickoff, monthly mentorship clinics, and leadership analytics. Lifetime updates as our client work evolves. Scale Hundreds to thousands of seats Covers The complete curriculum, self-paced with role paths Leaves with One shared foundation and progress analytics Per seat · volume bands Discuss seats → At 1,000+ developers: residency seed → 20–40-seat pilot → waves of 50–250 in parallel streams → top graduates co-instruct as internal champions. Per-developer cost drops as the programme scales. A Python foundations route exists for mixed-skill cohorts. 02 The system", + "heading": "Enterprise Academy licence", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-39bac4d3f6b961ad81494fbd", + "text": "One to six future AI leaders work with our senior architects: weekly coaching, applied labs, architecture reviews, and a capstone from a real internal workflow." + }, + { + "span_id": "span-33e33b6162cae1768604f275", + "text": "Each resident turns one workflow into a reference build later cohorts train on." + }, + { + "span_id": "span-61209122b2e6789fdf11491e", + "text": "Scale 1 to 6 senior engineers per wave Covers Workflow qualification, baselines, adoption, the demo-to-production gap Leaves with Reference builds, internal instructors, your standards" + }, + { + "span_id": "span-7b321d4084ca8d5bb5ddd7a6", + "text": "Per resident · priced as senior design work Nominate residents →" + }, + { + "span_id": "span-c6d8be017faa022d8dbf9bf9", + "text": "Per seat" + }, + { + "span_id": "span-41f91023c40231e296df4ae5", + "text": "The full Academy bundle: Full Stack AI Engineering (92 lessons), Agentic AI Engineering (5.0-rated), the Agentic Development course, and AI for Work, with live kickoff, monthly mentorship clinics, and leadership analytics." + }, + { + "span_id": "span-82a2a6dc804dd696757a16aa", + "text": "Lifetime updates as our client work evolves." + }, + { + "span_id": "span-e94207e09461bf5d2262b255", + "text": "Scale Hundreds to thousands of seats Covers The complete curriculum, self-paced with role paths Leaves with One shared foundation and progress analytics" + }, + { + "span_id": "span-3074ddbf1308e5b9ac9465cc", + "text": "Per seat · volume bands Discuss seats →" + }, + { + "span_id": "span-872c67a62eb67c1f2010816f", + "text": "At 1,000+ developers: residency seed → 20–40-seat pilot → waves of 50–250 in parallel streams → top graduates co-instruct as internal champions." + }, + { + "span_id": "span-b996d8832788c25e26a2a742", + "text": "Per-developer cost drops as the programme scales." + }, + { + "span_id": "span-a1c9a70fad7dbec8cb263829", + "text": "A Python foundations route exists for mixed-skill cohorts." + }, + { + "span_id": "span-a0e6194fd657852a3800e418", + "text": "02 The system" + } + ] + }, + { + "chunk_id": "tai-c626df4824ae3960f6e571da", + "text": "Two steps: use AI to build software, then build AI into it\nStep 1 makes every developer effective with coding agents, and its debrief names who belongs here. You are on the Step 2 page. Step 1 · Agentic developers Step 2 · AI engineers ← this page Who Every software developer, no AI background needed Selected builders , typically the top 10–20% They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems Outcome Ship faster on today's roadmap The capability you cannot hire : AI products and systems, built in-house Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · 8–12-week residency · per-seat licence When Now: it pays back on work already in flight After Step 1, or now if your devs already work seriously with agents Not there yet? Start with Step 1 → Same instructors, same standard, one scoping call covers both. 03 Why convert", + "heading": "Two steps: use AI to build software, then build AI into it", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-7d0bf5fc55fbaf431d333f9e", + "text": "Step 1 makes every developer effective with coding agents, and its debrief names who belongs here." + }, + { + "span_id": "span-829b5ec7a94406e85035a5b9", + "text": "You are on the Step 2 page." + }, + { + "span_id": "span-ea7673da023b1d665548fc14", + "text": "Who Every software developer, no AI background needed Selected builders , typically the top 10–20%" + }, + { + "span_id": "span-820c220755e0d3ce4f345154", + "text": "They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems" + }, + { + "span_id": "span-68a497ddb15fca4354ee6dfb", + "text": "Outcome Ship faster on today's roadmap The capability you cannot hire : AI products and systems, built in-house" + }, + { + "span_id": "span-81fadc46631d6daead992911", + "text": "Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · 8–12-week residency · per-seat licence" + }, + { + "span_id": "span-3d2d89a7b5c277eeb1da7a2d", + "text": "When Now: it pays back on work already in flight After Step 1, or now if your devs already work seriously with agents" + }, + { + "span_id": "span-c5a95fcdb8c55ce236ba23c5", + "text": "Not there yet?" + }, + { + "span_id": "span-b5cdfb98071d020555bf572a", + "text": "Start with Step 1 → Same instructors, same standard, one scoping call covers both." + }, + { + "span_id": "span-5e982b4d56e5514d2b79566d", + "text": "03 Why convert" + } + ] + }, + { + "chunk_id": "tai-256fd59a96518a1ad61f56e9", + "text": "Deployed AI pays back. Pilots without owners don't.\nThe pilots graveyard MIT's 2025 study attributes the failure of ~95% of enterprise GenAI pilots to a “learning gap”: workflow and people integration, not model quality. S&P Global finds 42% of enterprises abandoned most AI initiatives in 2025, up from 17%. Pilots stall because nobody owns production: design, evaluation, deployment, operations. MIT NANDA 2025 · S&P Global 2025 Deployment pays McKinsey finds only 7% of firms have deployed AI organisation-wide while 65% are stuck at proof of concept, yet 95% of deployed AI meets or exceeds ROI expectations, and 94% of firms are increasing spend into exactly that gap. The missing input isn't budget or models. It's people who can carry systems to production. McKinsey 2026 You can't hire your way out AI engineer is the #1 fastest-growing US job, postings up 143% year on year, with a 56% wage premium for AI skills, and LinkedIn's own data shows the most common feeder role is the software engineer. Your conversion pipeline is already on payroll; a new hire spends a year learning what your team already knows. LinkedIn 2026 · PwC 2025 Capability compounds The first trained cohort sets the engineering standard every later cohort inherits. Graduates seed later waves as internal champions and co-instructors, per-developer cost drops as the programme scales, and the capability stays in-house, with no dependence on us for the next use case. Programme design · scale path above 04 Straight answers", + "heading": "Deployed AI pays back. Pilots without owners don't.", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-958482a61b8a11ce45d2e1d4", + "text": "The pilots graveyard MIT's 2025 study attributes the failure of ~95% of enterprise GenAI pilots to a “learning gap”: workflow and people integration, not model quality." + }, + { + "span_id": "span-de8a9466d3e6f90771ab9e76", + "text": "S&P Global finds 42% of enterprises abandoned most AI initiatives in 2025, up from 17%." + }, + { + "span_id": "span-71c1acbefd49640d15e8dc3e", + "text": "Pilots stall because nobody owns production: design, evaluation, deployment, operations." + }, + { + "span_id": "span-2fd91aa8f2155a4493bbc372", + "text": "MIT NANDA 2025 · S&P Global 2025" + }, + { + "span_id": "span-7da132d11aaeedfedc699973", + "text": "Deployment pays McKinsey finds only 7% of firms have deployed AI organisation-wide while 65% are stuck at proof of concept, yet 95% of deployed AI meets or exceeds ROI expectations, and 94% of firms are increasing spend into exactly that gap." + }, + { + "span_id": "span-b967e8e1c2cbfb71e87c5d72", + "text": "The missing input isn't budget or models." + }, + { + "span_id": "span-057f8c334a48a6826cf5d294", + "text": "It's people who can carry systems to production." + }, + { + "span_id": "span-437b0ad69357dd05c098da49", + "text": "McKinsey 2026" + }, + { + "span_id": "span-e7ee5fd1bf1f7010a917bf0d", + "text": "You can't hire your way out AI engineer is the #1 fastest-growing US job, postings up 143% year on year, with a 56% wage premium for AI skills, and LinkedIn's own data shows the most common feeder role is the software engineer." + }, + { + "span_id": "span-989d6bcc14e24b7d751b64a1", + "text": "Your conversion pipeline is already on payroll; a new hire spends a year learning what your team already knows." + }, + { + "span_id": "span-2f05ad302bb658c0706331ff", + "text": "LinkedIn 2026 · PwC 2025" + }, + { + "span_id": "span-f3781897c4f2fdcb7f22b284", + "text": "Capability compounds The first trained cohort sets the engineering standard every later cohort inherits." + }, + { + "span_id": "span-2b439ac27bafa1dd9e05da98", + "text": "Graduates seed later waves as internal champions and co-instructors, per-developer cost drops as the programme scales, and the capability stays in-house, with no dependence on us for the next use case." + }, + { + "span_id": "span-4b96ad4cd321e1c65a8a4e00", + "text": "Programme design · scale path above" + }, + { + "span_id": "span-ed8aac59ae4af3c3cd303f3c", + "text": "04 Straight answers" + } + ] + }, + { + "chunk_id": "tai-44a349a825ad8c4357a3b768", + "text": "The questions your board will ask\n“Why not just hire AI engineers?” The role is barely three years old, supply is thin, the labs hire each other first, and frontier comp is extreme. Meanwhile your developers already hold the domain, systems, and codebase knowledge AI systems run on. Conversion wins on cost, speed, retention, and context. “Most AI initiatives fail anyway.” They fail for skills reasons: MIT attributes the 95% pilot-failure rate to a learning gap, not model quality, while 95% of deployed AI meets or exceeds ROI expectations (McKinsey). Capability-building is precisely the variable that moves you across that divide. “Training doesn't stick.” Lecture-style training doesn't. This programme certifies against shipped capstones (a deployed, evaluated LLM system per engineer) with exercises drawn from your approved roadmap, and closes with a written capability debrief. Nobody is certified for attending. “We can't take engineers off delivery.” Conversion runs part-time around delivery : self-paced study plus scheduled live checkpoints, shaped with engineering leadership. The residency is a defined weekly commitment for one to six seniors, not weeks out of the roadmap. “How do we choose who to convert?” Run Step 1 first: its capability debrief names the builders ready for this programme. Otherwise we baseline and segment the population in scoping (developers, data scientists, existing agent users) and agree the first cohort with engineering leadership. “Generic courses won't fit our stack.” Depth and emphasis are agreed against your engineers' baseline; electives are added where your stack needs them; exercises come from your roadmap, never generic demos. Every pattern we teach appears in a system we built ; see below. 05 Full transparency", + "heading": "The questions your board will ask", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-c941919847e8d99e7cd36b12", + "text": "“Why not just hire AI engineers?” The role is barely three years old, supply is thin, the labs hire each other first, and frontier comp is extreme." + }, + { + "span_id": "span-685663d10e48099ee9b8a6a3", + "text": "Meanwhile your developers already hold the domain, systems, and codebase knowledge AI systems run on." + }, + { + "span_id": "span-9815b1f30ddf9811c08db785", + "text": "Conversion wins on cost, speed, retention, and context." + }, + { + "span_id": "span-b231ad9d18bb33a289563b99", + "text": "“Most AI initiatives fail anyway.” They fail for skills reasons: MIT attributes the 95% pilot-failure rate to a learning gap, not model quality, while 95% of deployed AI meets or exceeds ROI expectations (McKinsey)." + }, + { + "span_id": "span-05ca1574e5ce820896179a7b", + "text": "Capability-building is precisely the variable that moves you across that divide." + }, + { + "span_id": "span-d8fbc729df7ca2f136d6b81c", + "text": "“Training doesn't stick.” Lecture-style training doesn't." + }, + { + "span_id": "span-7178387a412d655330dab45b", + "text": "This programme certifies against shipped capstones (a deployed, evaluated LLM system per engineer) with exercises drawn from your approved roadmap, and closes with a written capability debrief." + }, + { + "span_id": "span-f09df88d17476ad1a39d4f5c", + "text": "Nobody is certified for attending." + }, + { + "span_id": "span-2b8ffda616aa3015baea2bcd", + "text": "“We can't take engineers off delivery.” Conversion runs part-time around delivery : self-paced study plus scheduled live checkpoints, shaped with engineering leadership." + }, + { + "span_id": "span-3ea7c22307c1eec8a1d3b816", + "text": "The residency is a defined weekly commitment for one to six seniors, not weeks out of the roadmap." + }, + { + "span_id": "span-e2a41a07020755665eb9805c", + "text": "“How do we choose who to convert?” Run Step 1 first: its capability debrief names the builders ready for this programme." + }, + { + "span_id": "span-8f693809796902ae382db4cc", + "text": "Otherwise we baseline and segment the population in scoping (developers, data scientists, existing agent users) and agree the first cohort with engineering leadership." + }, + { + "span_id": "span-76b96a974262ecdc6cc28414", + "text": "“Generic courses won't fit our stack.” Depth and emphasis are agreed against your engineers' baseline; electives are added where your stack needs them; exercises come from your roadmap, never generic demos." + }, + { + "span_id": "span-d56240ec9456c478e2a65683", + "text": "Every pattern we teach appears in a system we built ; see below." + }, + { + "span_id": "span-246c6b9e85fac759f47c5259", + "text": "05 Full transparency" + } + ] + }, + { + "chunk_id": "tai-dc2648de7d06bb81b2740344", + "text": "The curriculum , module by module\nDepth and emphasis are agreed against your engineers' baseline in scoping. Exercises come from your approved roadmap, never generic demos. 01 Foundations and model behaviour core + 1 Models, prompting, and the limitations that shape engineering decisions 2 Security, prompt injection, and responsible technical use 3 Choosing the right technique for the task: prompting, RAG, fine-tuning, or agents 02 Systems: retrieval and application patterns build + 1 Retrieval-augmented generation, from first pipeline to production patterns 2 Framework and architecture trade-offs; context and application design 3 Application-oriented exercises against approved use cases 03 Evaluation and reliability rigour + 1 Test-set design and evaluation as an engineering discipline 2 Generation and retrieval evaluation; LLM-as-judge and regression thinking 3 Observability, so output is never accepted on intuition alone 04 Agent engineering advanced + 1 Agentic patterns and when agents beat simpler techniques; tool use and function calling 2 Orchestration and multi-agent systems; MCP integrations; memory and state design 3 Agent evaluation, guardrails, and human-review boundaries 05 Optimisation and deployment production + 1 Fine-tuning and optimisation; deployment and cost control 2 Monitoring in production, and staying current as the field moves 3 Electives where your stack needs them: large codebases, domain integrations, compliance 06 Scoped to your baseline custom + 1 Depth and emphasis agreed against the engineers' baseline during scoping 2 Module weighting follows your use cases, not a fixed timetable 3 No stage is filler for the next 07 Capstone: certified against shipped work certified + 1 Every module lands in capstone work: a deployed, evaluated LLM", + "heading": "The curriculum , module by module", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-9005a858d7b42048087d5275", + "text": "Depth and emphasis are agreed against your engineers' baseline in scoping." + }, + { + "span_id": "span-c445c8295fc280e7e5870a04", + "text": "Exercises come from your approved roadmap, never generic demos." + }, + { + "span_id": "span-d3ebfeb6ee8ee50845f698da", + "text": "01 Foundations and model behaviour core + 1 Models, prompting, and the limitations that shape engineering decisions" + }, + { + "span_id": "span-b9f479a7adbadfa7ca7f5bce", + "text": "2 Security, prompt injection, and responsible technical use" + }, + { + "span_id": "span-938c62b1f86dd2f0d2e34d41", + "text": "3 Choosing the right technique for the task: prompting, RAG, fine-tuning, or agents" + }, + { + "span_id": "span-dc5f9f8a34ee128754e96190", + "text": "02 Systems: retrieval and application patterns build + 1 Retrieval-augmented generation, from first pipeline to production patterns" + }, + { + "span_id": "span-1c1f3d5f45abcd2c3f7906a5", + "text": "2 Framework and architecture trade-offs; context and application design" + }, + { + "span_id": "span-6dd8e427e1ba048395adfba7", + "text": "3 Application-oriented exercises against approved use cases" + }, + { + "span_id": "span-c371034230206b5ea553ae43", + "text": "03 Evaluation and reliability rigour + 1 Test-set design and evaluation as an engineering discipline" + }, + { + "span_id": "span-09a447e70b68fbc48fdf5c04", + "text": "2 Generation and retrieval evaluation; LLM-as-judge and regression thinking" + }, + { + "span_id": "span-9603bfde44bc304950c4c700", + "text": "3 Observability, so output is never accepted on intuition alone" + }, + { + "span_id": "span-a74615a04553324ceb055d36", + "text": "04 Agent engineering advanced + 1 Agentic patterns and when agents beat simpler techniques; tool use and function calling" + }, + { + "span_id": "span-05539575cf5b9c8613b11fef", + "text": "2 Orchestration and multi-agent systems; MCP integrations; memory and state design" + }, + { + "span_id": "span-17d6aca3ab082e8ee58eedce", + "text": "3 Agent evaluation, guardrails, and human-review boundaries" + }, + { + "span_id": "span-45897df9b38a0073735639ee", + "text": "05 Optimisation and deployment production + 1 Fine-tuning and optimisation; deployment and cost control" + }, + { + "span_id": "span-7ff2f672d09e3a1bfd359f7b", + "text": "2 Monitoring in production, and staying current as the field moves" + }, + { + "span_id": "span-6071d68a340bca65c8c151c6", + "text": "3 Electives where your stack needs them: large codebases, domain integrations, compliance" + }, + { + "span_id": "span-8f8209c4b8c49bb3e3f7179a", + "text": "06 Scoped to your baseline custom + 1 Depth and emphasis agreed against the engineers' baseline during scoping" + }, + { + "span_id": "span-b54351c88f23666e420319bf", + "text": "2 Module weighting follows your use cases, not a fixed timetable" + }, + { + "span_id": "span-2590b7ea3410286298befea9", + "text": "3 No stage is filler for the next" + } + ] + }, + { + "chunk_id": "tai-877c3c8261133687d11f505f", + "text": "The curriculum , module by module\nsystem 2 Certification against shipped capstones, not attendance 3 A written capability debrief closes each cohort Outcome: engineers who design, evaluate, ship, and maintain production LLM systems, certified against a deployed capstone. 06 We teach what we ship", + "heading": "The curriculum , module by module", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-029bc7555de4c0882d0a91d5", + "text": "2 Certification against shipped capstones, not attendance" + }, + { + "span_id": "span-c4d4a42e1213ad6fbe585d19", + "text": "3 A written capability debrief closes each cohort" + }, + { + "span_id": "span-bb62729e11f0a2b713b4f33c", + "text": "Outcome: engineers who design, evaluate, ship, and maintain production LLM systems, certified against a deployed capstone." + }, + { + "span_id": "span-62a99418f200dd0e05ac271c", + "text": "06 We teach what we ship" + } + ] + }, + { + "chunk_id": "tai-b85906c4fbbf19f294aee0d2", + "text": "Every pattern we teach appears in a system we built\n21+ AI applications shipped to production across finance, government, law enforcement, healthcare, and industry. The curriculum is drawn from that work, which is why your training partner must be a practising builder. Banker-reasoning engine Banking · Europe Agentic financial-intelligence platform routing 1M+ words per company across four frontier model families; ~800 prompts and 20-step workflows per analysis. Teaches: routing, orchestration, and evaluation, from idea to live demos with global banks. Dialog Insight Martech · Canada Multi-month production-agent enablement: weekly advisory, 16+ hours of guided code transfer, six runnable reference repositories, a 25-scenario benchmark. Outcome: working code the team owns: a production-shaped agent stack, runnable and auditable. Document extraction at scale Mortgage · Canada 50+ bilingual document types for underwriting; three independent extraction passes with consensus; never-overwrite Salesforce integration; data-residency compliant. Teaches: structured outputs, consensus architectures, human-review boundaries. “The most comprehensive textbook to date on building LLM applications; every essential topic in an AI engineer's toolkit.” Jerry Liu Founder & CEO, LlamaIndex, on our book “An indispensable guide for anyone venturing into the world of LLMs, covering everything from theory to practical deployment.” Senior Data Engineer Meta “Invaluable to anyone looking to dive into the field quickly and efficiently.” Senior Applied Research Scientist Mila (Yoshua Bengio's AI lab) 07 Proven on ourselves", + "heading": "Every pattern we teach appears in a system we built", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-fbc4e68ce4a1a01cb7e8c7e9", + "text": "21+ AI applications shipped to production across finance, government, law enforcement, healthcare, and industry." + }, + { + "span_id": "span-73a264f9cd78695f173eb999", + "text": "The curriculum is drawn from that work, which is why your training partner must be a practising builder." + }, + { + "span_id": "span-c388565b910ccef2ae24f7eb", + "text": "Banker-reasoning engine Banking · Europe" + }, + { + "span_id": "span-57e9c7b792db7614370a82f5", + "text": "Agentic financial-intelligence platform routing 1M+ words per company across four frontier model families; ~800 prompts and 20-step workflows per analysis." + }, + { + "span_id": "span-209b5bd5ed619fd9afb1720e", + "text": "Teaches: routing, orchestration, and evaluation, from idea to live demos with global banks." + }, + { + "span_id": "span-14281d84eb20a5fd501c37bc", + "text": "Dialog Insight Martech · Canada" + }, + { + "span_id": "span-0f152c01ab6813e064a7cdf6", + "text": "Multi-month production-agent enablement: weekly advisory, 16+ hours of guided code transfer, six runnable reference repositories, a 25-scenario benchmark." + }, + { + "span_id": "span-6740f9d0bfa103e4ad9ee591", + "text": "Outcome: working code the team owns: a production-shaped agent stack, runnable and auditable." + }, + { + "span_id": "span-f01c49078fe2d6d3f0d20127", + "text": "Document extraction at scale Mortgage · Canada" + }, + { + "span_id": "span-2a3432bc2911be14fdcddbfc", + "text": "50+ bilingual document types for underwriting; three independent extraction passes with consensus; never-overwrite Salesforce integration; data-residency compliant." + }, + { + "span_id": "span-2a75683080fa1704748950e8", + "text": "Teaches: structured outputs, consensus architectures, human-review boundaries." + }, + { + "span_id": "span-49eaef208bf6ea737ab20b25", + "text": "“The most comprehensive textbook to date on building LLM applications; every essential topic in an AI engineer's toolkit.”" + }, + { + "span_id": "span-d87c95c03d7498a0e1078195", + "text": "Jerry Liu" + }, + { + "span_id": "span-a1a80d25bae59bf5a2fbbf7b", + "text": "Founder & CEO, LlamaIndex, on our book" + }, + { + "span_id": "span-3179f933727b2ed85c889aed", + "text": "“An indispensable guide for anyone venturing into the world of LLMs, covering everything from theory to practical deployment.”" + }, + { + "span_id": "span-7e69d4fa3188234084520235", + "text": "Senior Data Engineer" + }, + { + "span_id": "span-35cad011772471428f47420a", + "text": "“Invaluable to anyone looking to dive into the field quickly and efficiently.”" + }, + { + "span_id": "span-e7eee923fe2e3f8fbf9902e3", + "text": "Senior Applied Research Scientist" + }, + { + "span_id": "span-806c7b21126434a002d174b3", + "text": "Mila (Yoshua Bengio's AI lab)" + }, + { + "span_id": "span-6ab81c0883623eebbcada51d", + "text": "07 Proven on ourselves" + } + ] + }, + { + "chunk_id": "tai-38d372e68e52fe2fc7fe3474", + "text": "We ran this conversion line on ourselves first\nOur delivery arm needed more production AI engineers than the market could supply, so we built a conversion line: Learner → Author → Instructor → Engineer. 15+ of our own engineers came off it. Work is the audition, and your cohorts are held to the same standard. LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production (10k+ copies). Chief AI Officer for a banking agentic-engine build. Ex-J.P. Morgan VP, Research. LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA (Yoshua Bengio's lab). Built our 100k+ member community. FC Fabio Chiusano Senior AI Engineer NLP lead; ex-founder & Head of Data Science. Joined as a contributing author; proof the pipeline works. AK Assel Kashkenbayeva Deployment lead Ex-Palantir project lead; UCL PhD. Carries the forward-deployment model the residency teaches. 500k+ AI practitioners taught since 2019 15+ in-house AI engineers, grown from our own line 21+ AI applications shipped to production 10k+ copies of Building LLMs for Production sold 08 How it runs\n\nEvidence before scale\n01\n\nScoping call · 30–60 min\nTeam sizes, stacks, baseline, and target capabilities, discussed with the instructors. We'll tell you if Step 1 should come first. 02\n\nBaseline & co-design\nWe assess and segment the developer population; a nominee per team keeps materials resonant with real work. Agreed with engineering leadership before anything is booked. 03\n\nPilot cohort\nTypically 20 to 40 engineers, chosen with leadership. Exercises from your approved roadmap, never generic demos. Capstones build throughout. 04", + "heading": "We ran this conversion line on ourselves first", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-ecbf0f9226b7c0f5e0a8149b", + "text": "Our delivery arm needed more production AI engineers than the market could supply, so we built a conversion line: Learner → Author → Instructor → Engineer." + }, + { + "span_id": "span-37cbbbe58734357eab5374d0", + "text": "15+ of our own engineers came off it." + }, + { + "span_id": "span-c51e77b0a893ea098634d216", + "text": "Work is the audition, and your cohorts are held to the same standard." + }, + { + "span_id": "span-4dc54c58808260f242d14402", + "text": "LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production (10k+ copies)." + }, + { + "span_id": "span-2901ed890b88945fbaacdbc2", + "text": "Chief AI Officer for a banking agentic-engine build." + }, + { + "span_id": "span-2e1aac8ee282ca31bf2bbffb", + "text": "Ex-J.P." + }, + { + "span_id": "span-ac20f34421acb21d8dba3a01", + "text": "Morgan VP, Research." + }, + { + "span_id": "span-234bd5ed4af360606040c8c0", + "text": "LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA (Yoshua Bengio's lab)." + }, + { + "span_id": "span-80d1f2ca7372323200c43c10", + "text": "Built our 100k+ member community." + }, + { + "span_id": "span-c5c4a4275a9fd3fcbcaf8fb1", + "text": "FC Fabio Chiusano Senior AI Engineer NLP lead; ex-founder & Head of Data Science." + }, + { + "span_id": "span-6531b902a2f99df43bcdb8a3", + "text": "Joined as a contributing author; proof the pipeline works." + }, + { + "span_id": "span-5470633e608df9c54a17025d", + "text": "AK Assel Kashkenbayeva Deployment lead Ex-Palantir project lead; UCL PhD." + }, + { + "span_id": "span-a9fe1e63b1f5cefcb4b8dff4", + "text": "Carries the forward-deployment model the residency teaches." + }, + { + "span_id": "span-21bf779fa7ddf8619a55bb5f", + "text": "500k+ AI practitioners taught since 2019" + }, + { + "span_id": "span-3fccd9d11a273c9ee9ff69db", + "text": "15+ in-house AI engineers, grown from our own line" + }, + { + "span_id": "span-00a67ed1b67c7e349d2752db", + "text": "21+ AI applications shipped to production" + }, + { + "span_id": "span-3a4ec0c2a3d3e050710b9e59", + "text": "10k+ copies of Building LLMs for Production sold" + }, + { + "span_id": "span-502f21d183aafbe3c4a8fa51", + "text": "08 How it runs" + }, + { + "span_id": "span-c2aad9c213d5d2fd0f48ce05", + "text": "Evidence before scale" + }, + { + "span_id": "span-6d8851e6ce5f30520b1fb921", + "text": "Scoping call · 30–60 min" + }, + { + "span_id": "span-761d84647120f8435fb09a7a", + "text": "Team sizes, stacks, baseline, and target capabilities, discussed with the instructors." + }, + { + "span_id": "span-7917787ac03406c4523f3860", + "text": "We'll tell you if Step 1 should come first." + }, + { + "span_id": "span-90f88442779a4f1d60b2d237", + "text": "We assess and segment the developer population; a nominee per team keeps materials resonant with real work." + }, + { + "span_id": "span-aeaec8066572123130f9e4cd", + "text": "Agreed with engineering leadership before anything is booked." + }, + { + "span_id": "span-748c9f55bb0cf031e5b92260", + "text": "Typically 20 to 40 engineers, chosen with leadership." + }, + { + "span_id": "span-5e6b02c2277efa946548d3e9", + "text": "Exercises from your approved roadmap, never generic demos." + }, + { + "span_id": "span-b7969e7bb0e17d420a4a2bcc", + "text": "Capstones build throughout." + } + ] + }, + { + "chunk_id": "tai-1f6fda503f40ac2e3c43123b", + "text": "Certify & scale decision\nCertification against shipped capstones and a written debrief of what the cohort can now do. We scale to further waves only when the debrief earns it. + Pilot first. One cohort, agreed success measures, and a scale-decision date, before any programme commitment. + You own the artefacts. Capstones, standards, and reference builds stay in-house. No dependence on us for the next use case. + Procurement-ready. NDAs standard, PO and invoicing supported, no production access required; your code stays yours. + Portfolio roll-outs. For groups and PE portfolios: shared standards, per-company scoping, champions carried across companies.\n\nThe winners of the next three years convert their existing developer base fastest.\nOne call, one plan, one pilot. Scale only when the debrief earns it. Book a scoping call Start with Step 1 09 Logistics", + "heading": "Certify & scale decision", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-2ad9d70b8732c904207513d0", + "text": "Certification against shipped capstones and a written debrief of what the cohort can now do." + }, + { + "span_id": "span-9e53ec3720162c1ba45fd56e", + "text": "We scale to further waves only when the debrief earns it." + }, + { + "span_id": "span-cd364d19db88bd5fd8ddd352", + "text": "+ Pilot first." + }, + { + "span_id": "span-3b4bfa3c367dcb946e1e6b14", + "text": "One cohort, agreed success measures, and a scale-decision date, before any programme commitment." + }, + { + "span_id": "span-0a07a05fee23b04a078239b7", + "text": "+ You own the artefacts." + }, + { + "span_id": "span-e3a3afc00158d75fa217990e", + "text": "Capstones, standards, and reference builds stay in-house." + }, + { + "span_id": "span-18654eefa1e4e8e259d4861b", + "text": "No dependence on us for the next use case." + }, + { + "span_id": "span-c1c7e550d1d93c817803fc6b", + "text": "+ Procurement-ready." + }, + { + "span_id": "span-a978b99e56629bc9cb29bbfd", + "text": "NDAs standard, PO and invoicing supported, no production access required; your code stays yours." + }, + { + "span_id": "span-2c59473bac029d2d1e185a2a", + "text": "+ Portfolio roll-outs." + }, + { + "span_id": "span-5f9ab45ce6208d919c59f99a", + "text": "For groups and PE portfolios: shared standards, per-company scoping, champions carried across companies." + }, + { + "span_id": "span-056112bb51820104d8bab07d", + "text": "One call, one plan, one pilot." + }, + { + "span_id": "span-2de135f53e646711fd375566", + "text": "Scale only when the debrief earns it." + }, + { + "span_id": "span-f9ce77ddab755623280a292e", + "text": "Book a scoping call Start with Step 1" + }, + { + "span_id": "span-264fe9eef43055f586c040fe", + "text": "09 Logistics" + } + ] + }, + { + "chunk_id": "tai-b177612fd6a839c6ceee1cca", + "text": "Frequently asked questions\nDo participants need to do Step 1 first? + It's the recommended order, not a hard rule. Step 1 builds the agent fluency this programme assumes, and its debrief tells you who to send here. If your developers already work seriously with coding agents, we confirm the baseline in scoping and start directly. Hard prerequisites: working Python and software engineering experience; a Python foundations route exists for mixed cohorts. Where do capstones fit if delivery is part-time? + Cohorts run over weeks around delivery: self-paced study, live checkpoints, and capstone work that builds throughout. Certification happens when the capstone ships (a deployed, evaluated system), not when the calendar runs out. Does \"conversion\" mean a guaranteed role change? + No. The programme develops applied AI engineering capability in existing software developers, certified against shipped capstone work. Scope, assessment, and outcomes are agreed per engagement. What proof can we review before buying? + The curriculum is on this page, the underlying courses have public previews, our book is in print, and the production systems behind the curriculum are described above. References and the most relevant engagements are discussed in scoping. 10 Next step", + "heading": "Frequently asked questions", + "index": 10, + "evidence_spans": [ + { + "span_id": "span-3fe0ad45dbb5ef8d7e06e79d", + "text": "Cohorts run over weeks around delivery: self-paced study, live checkpoints, and capstone work that builds throughout." + }, + { + "span_id": "span-49814bb3626f6caac2c476fc", + "text": "Certification happens when the capstone ships (a deployed, evaluated system), not when the calendar runs out." + }, + { + "span_id": "span-c37b86631e899079773dd9a4", + "text": "The programme develops applied AI engineering capability in existing software developers, certified against shipped capstone work." + }, + { + "span_id": "span-b109e2b76456b5890e46ce57", + "text": "Scope, assessment, and outcomes are agreed per engagement." + }, + { + "span_id": "span-47571d9da92cf840dedfa8fb", + "text": "The curriculum is on this page, the underlying courses have public previews, our book is in print, and the production systems behind the curriculum are described above." + }, + { + "span_id": "span-0d7864481513c938fdfbb69f", + "text": "References and the most relevant engagements are discussed in scoping." + }, + { + "span_id": "span-66e67e95e652fe8c254f7c80", + "text": "10 Next step" + } + ] + }, + { + "chunk_id": "tai-549d76502a8283dfa1721144", + "text": "Book a scoping call\nTell us about your engineering team, current baseline, and the AI capabilities you want to build. We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the programme. 1 On the call: team sizes, stacks, baseline, target capabilities, and which formats fit which teams. 30–60 minutes, with instructors. 2 You leave with: a recommended programme architecture, workload estimate, commercial range, and measurement plan. 3 No pressure: if Step 1, or nothing, is the right answer, we say so. Every enquiry is read by the Towards AI team. Prefer email? louis@towardsai.net Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet AI engineer conversion cohort AI leader residency Enterprise Academy licence Step 1: Agentic developer programme What does your team need? I agree to be contacted about this request. Book a scoping call", + "heading": "Book a scoping call", + "index": 11, + "evidence_spans": [ + { + "span_id": "span-e2901c3e629b60a8ab600463", + "text": "Tell us about your engineering team, current baseline, and the AI capabilities you want to build." + }, + { + "span_id": "span-2aecf2b7b0b248025f10349b", + "text": "We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the programme." + }, + { + "span_id": "span-9f4fd3e54c5e83543e505c92", + "text": "1 On the call: team sizes, stacks, baseline, target capabilities, and which formats fit which teams." + }, + { + "span_id": "span-a4396ad6259cc5df97a21c1f", + "text": "30–60 minutes, with instructors." + }, + { + "span_id": "span-406f187a29f48c40a36137a4", + "text": "2 You leave with: a recommended programme architecture, workload estimate, commercial range, and measurement plan." + }, + { + "span_id": "span-8635df54c7bbce5dc2f111bc", + "text": "3 No pressure: if Step 1, or nothing, is the right answer, we say so." + }, + { + "span_id": "span-4a1adacad0b35720c1d28e9c", + "text": "Every enquiry is read by the Towards AI team." + }, + { + "span_id": "span-130782821792f3ee8bff7811", + "text": "Prefer email? louis@towardsai.net" + }, + { + "span_id": "span-4f080f52bf8f1fb312060c1b", + "text": "Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet AI engineer conversion cohort AI leader residency Enterprise Academy licence Step 1: Agentic developer programme What does your team need?" + }, + { + "span_id": "span-35cf4e1b0595adb8af65dd57", + "text": "I agree to be contacted about this request." + }, + { + "span_id": "span-a1f5528e01b6e2348af53c85", + "text": "Book a scoping call" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "77b1703a94995f87969ef66284b1f289be822b4974c4cfb9110abd8843cd4f18", + "content_hash": "77b1703a94995f87969ef66284b1f289be822b4974c4cfb9110abd8843cd4f18", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/enterprise/software-developer-to-ai-engineer/" + ], + "evidence_hash": "002a09b3cb893f88958a64d15c405cbed31d3f38a4b1ed928412d0fdad699a3d" + }, + { + "discovered_url": "https://towardsai.com/enterprise/agentic-developer-conversion/", + "url": "https://towardsai.com/enterprise/agentic-developer-conversion/", + "canonical_url": "https://towardsai.com/enterprise/agentic-developer-conversion/", + "host": "towardsai.com", + "path": "/enterprise/agentic-developer-conversion", + "kind": "b2b", + "offer_id": null, + "entity_id": "page:towardsai.com/enterprise/agentic-developer-conversion", + "title": "Claude Code & Codex Team Training: One-Day Agentic Developer Bootcamp | Towards AI", + "review_title": "Agentic Developer Conversion", + "reviewed_summary": "Enterprise coding-agent adoption program.", + "meta_description": "Your team has coding-agent seats; we install the discipline that makes them pay. Start with a one-day agentic developer bootcamp for up to 40 developers, taught by engineers who ship with these tools daily. Book a scoping call.", + "headings": [ + "Your team has coding-agent seats. Now make them pay.", + "Pick your starting point", + "One-day agentic developer bootcamp", + "Deep-dive team bootcamp", + "Enterprise Academy licence", + "Two steps: use AI to build software, then build AI into it", + "Seats are everywhere. Skill is not.", + "The questions your CFO will ask", + "Trained where mistakes aren't an option", + "No training faculty. Practising engineers.", + "Low commitment in, evidence out", + "Scoping call · 30 min", + "Baseline", + "The day(s)", + "Debrief & decision", + "One day. Up to 40 developers. Your approved tools.", + "Frequently asked questions", + "Book a scoping call" + ], + "text": "Step 1 of 2 · Every developer\n\nYour team has coding-agent seats. Now make them pay.\nCoding agents amplify the discipline a team already has. We install that discipline in one day, inside your approved Claude Code or Codex setup. Start with one day Book a scoping call 30 minutes with the instructors who deliver it. No sales team. 8x more code merged per engineer at Anthropic since 2024, with trained agent use −19% experienced devs slower with AI when untrained, while believing they were faster (METR) 1 day to install the shared operating practice, on-site or remote, up to 40 developers Programme at a glance Start 1-day bootcamp , up to 40 developers Scale 2–3-day deep dive · per-seat licence Tools Claude Code · Codex, your approved setup Where On-site or remote, English or French Taught by Engineers who ship with these tools daily Then Debrief names your Step-2 AI engineer candidates Training delivered for teams at 01 Start here\n\nPick your starting point\nThree ways in, one operating standard. Most teams prove it in one day, then scale. Every bootcamp ends with a written capability debrief and our recommendation on whether to go further, including “don't.” 1 day Start here\n\nOne-day agentic developer bootcamp\nOne concentrated day inside your approved setup. The operating loop of plan, direct, review, and verify, plus repository instructions (CLAUDE.md / AGENTS.md) drafted against one of your real repos, and your team conventions v1. Cohort Up to 40 developers Leaves with Working setups, drafted agent files, conventions v1, debrief Best for Proving trained agent use to a sceptical team, fast Fixed fee per cohort · quoted at scoping Book the day → 2–3 days Most chosen\n\nDeep-dive team bootcamp\nTwo instructor-led days across all eight modules: agent files, plan mode, large codebases, multi-agent and MCP, guardrails, cost. Optional applied day on tasks from your own roadmap. Cohort 10 to 40 engineers; repeatable across teams Leaves with Team standards, review discipline, multi-agent practice, debrief Best for Teams defining shared conventions properly Per cohort · quoted after scoping Scope the bootcamp → Per seat\n\nEnterprise Academy licence\nThe full self-paced Academy bundle, including the Agentic Development course, with role-based paths, live kickoff, monthly mentorship clinics, office hours, and leadership analytics. The baseline for hundreds to thousands of developers. Scale Hundreds to thousands of seats Leaves with One shared foundation, progress analytics, lifetime updates Best for Consistent baseline across a large org Per seat · volume bands Discuss seats → Also available: 1:1 coaching for the lead writing your conventions. Cost is driven by participant count, live days, and customisation depth; an indicative quote accompanies the scoping agenda, before you commit to anything. 02 The system\n\nTwo steps: use AI to build software, then build AI into it\nStep 1 makes every developer effective with coding agents and reveals who should go further. Step 2 converts your strongest builders into AI engineers. You are on the Step 1 page. Step 1 · Agentic developers ← this page Step 2 · AI engineers Who Every software developer , no AI background needed Selected builders, typically the top 10–20% They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems Outcome Ship faster on today's roadmap ; agent code passes the same bar as human code New products and AI systems; the capability you cannot hire Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · senior residency · per-seat licence When Now: it pays back on work already in flight After Step 1: the debrief names who's ready Explore Step 2: AI engineer conversion → Same instructors, same standard, one scoping call covers both. 03 Why it works\n\nSeats are everywhere. Skill is not.\nYour developers already spend a median of two hours a day with AI. The field data is blunt about what separates the teams that win from the teams that stall. AI is an amplifier DORA's 2025 study of ~5,000 professionals: 90% of developers now use AI, a median of two hours a day, and over 80% report productivity gains, yet delivery instability rises unless the workflow is redesigned. Seven organisational capabilities decide whether individual gains ever reach the org. Amplifiers make disciplined teams faster and undisciplined teams slower. DORA 2025 · Google Cloud Untrained use backfires METR found experienced developers 19% slower with AI on familiar code, while believing they were 20% faster. Self-assessment is miscalibrated, so self-learning doesn't converge. 66% of developers fight “almost right” output; 45% say debugging AI code takes longer; only around one in six report significant productivity gains. METR 2025 · Stack Overflow 2025 The gap is trainable Inside Anthropic, Claude now authors 80%+ of merged code and code merged per engineer is up 8x since 2024. Experts drive ~12 agent actions and 3,200 words of output per prompt versus ~5 and 600 for novices. The difference is task framing, context, and verification: exactly what we train. BCG finds 5+ hours of structured training with coaching is what turns tool access into regular, effective use. Anthropic 2026 · BCG 2025 Waiting compounds the gap Google's CEO says AI now writes around three-quarters of the company's new code. Microsoft has told staff AI use is “no longer optional.” Mandates without enablement backfire; the winners build capability instead of compulsion, and the first trained cohort sets the standard every later cohort inherits. Alphabet 2026 · Microsoft 2025 04 Straight answers\n\nThe questions your CFO will ask\n“Our devs can self-learn from docs and YouTube.” They've had three years. Only ~16% of developers report significant productivity gains from AI, and METR shows devs believed they were 20% faster while measuring 19% slower: miscalibrated self-assessment means self-learning doesn't converge . Structured training with coaching is what moves usage (BCG 2025). “The tools change too fast; training goes stale.” The durable skills are tool-agnostic: planning, context engineering, review discipline, verification, team conventions. They transfer across Claude Code, Codex, and whatever ships next. We re-train on the frontier monthly because we ship with it , and Academy content carries lifetime updates. “We can't take engineers off delivery.” The entry format is one day . Your engineers already spend a median two hours a day with AI, unmanaged. The bootcamp redirects time already being spent, and applied exercises run on tasks from your own roadmap, not toy repos. “Security won't allow it.” Everything runs in your approved environment : your tool surface, your permissions, no production access required, NDAs standard. We've delivered inside Europol. And the biggest security risk is untrained use; review boundaries and guardrails are half the curriculum. “We already pay for seats, so we're covered.” Seats are access, not capability. DORA's finding is that AI amplifies the engineering discipline you already have ; licences amplify nothing on their own. The 8x gains show up where trained use meets shared standards. “How do we know it worked?” We capture a baseline in scoping and close with a written capability debrief : what changed in how your engineers work, the conventions your team keeps, and the metrics to watch: throughput, review load, spend per merged change. Scale to more cohorts only if the debrief earns it. 05 Delivered in the field\n\nTrained where mistakes aren't an option\nEuropol Law enforcement · EU Three-day security-first bootcamp: technical investigation teams went from cautious LLM users to confident builders of secure, reviewable intelligence pipelines. Outcome: special agents deploying AI across operational intelligence workflows. Major US public library system Public sector · USA Four-day, two-track programme for engineers and operations staff, backed by 43 hands-on notebooks and a hackathon on real pain points. Outcome: adoption from foundations to shipped build, owned by the library's own teams, not a vendor. Global banks & leading funds Finance Trader training at a leading global bank; live Claude Cowork sessions for investment teams at leading funds; enterprise bootcamps tailored to each firm's stack. Outcome: desk-level AI workflows in regulated, latency-sensitive environments. “Since the training, I’ve built two local apps using your vibe-coding approach. With every Gemini release I test vibe coding capabilities and have been unimpressed. Your approach is brilliant… We’re going to make it our primary developer model.” VP of IT Major US public library system “Excellent in-depth handling of trade-offs in evaluating and deploying agent-based solutions. I've used a lot of the advice in building our enterprise practices to agent use in my organisation.” Cathal Curtin Student, Agent Engineering 06 Who delivers\n\nNo training faculty. Practising engineers.\nThe people in the room ship client systems with coding agents every working week. The material is the conventions, failure modes, and review habits that survived real codebases. LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production. Newsletter read by 200k+ practitioners. Ex-J.P. Morgan VP. LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA, Yoshua Bengio's lab. FC Fabio Chiusano Senior AI Engineer NLP lead; ships the client systems the curriculum is drawn from. SV Samridhi Vaid Machine Learning Engineer Ships production multimodal AI, RAG, and agent systems; M.Sc. in Computing Science, University of Alberta. 500k+ AI practitioners taught since 2019 15+ in-house AI engineers shipping client systems 10k+ copies of Building LLMs for Production sold 21+ AI applications shipped to production 07 How it runs\n\nLow commitment in, evidence out\n01\n\nScoping call · 30 min\nTeams, approved tools, working patterns, and priorities, discussed with the instructors. You get a written plan and an indicative quote within one business day. 02\n\nBaseline\nWe capture where the team actually is (usage patterns, conventions, metrics worth watching) so the debrief measures change, not vibes. 03\n\nThe day(s)\nLive workshops inside your approved environment. No surprise tooling, no unapproved repository access, acceptance and merge stay with your engineers. 04\n\nDebrief & decision\nA written capability debrief: what changed, which conventions to keep, who's ready for Step 2, and whether further cohorts are worth it. Scale only if it earns it. + Start with one day. Prove it on one cohort before you commit to a programme. + Your tools, your rules. Scoped to the tool surface your organisation has approved. No production access required. + Procurement-ready. NDAs on request, PO and invoicing supported, references in scoping. + We'll say if it's not for you. If a cheaper format fits, or training isn't the right tool, the instructors tell you on the call.\n\nOne day. Up to 40 developers. Your approved tools.\nBook the scoping call; get the plan and quote within one business day. Book a scoping call 08 Logistics\n\nFrequently asked questions\nDoes the programme require one specific coding agent? + No. We scope workshops around the approved tool surface relevant to your team (Claude Code, Codex, or both) and explain differences where practices are tool-specific, including how to standardise across them. What exactly happens on the applied day? + Only with agreement: after sign-off on approved repositories or prepared tasks, permissions, and review expectations, the cohort works agreed tasks from your roadmap under instructor supervision. It's applied learning, never production delivery. One signature session: an hour where engineers may only plan, direct, and review while the agent writes everything; the diffs are then reviewed against your own bar. What are the prerequisites? + Software engineering experience and access to your approved coding-agent tooling. No AI background needed; that's the point of Step 1. What comes after? + The debrief names next steps: further cohorts, team conventions to keep, and the developers ready for Step 2: AI engineer conversion . Beyond training, the same engineers build custom AI systems and provide fractional AI leadership. 09 Next step\n\nBook a scoping call\nTell us about your developer cohorts, approved tools, and priorities. We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the workshops. 1 On the call: teams, stacks, baseline, and which format fits. 30 minutes, with instructors. 2 You leave with: a written plan, an indicative quote, and a measurement approach. 3 No pressure: if one day is enough, we say so. Every enquiry is read by the Towards AI team. Prefer email? louis@towardsai.net Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet One-day agentic bootcamp Deep-dive bootcamp (2-3 days) Enterprise Academy licence Leader coaching Step 2: AI engineer programme What should the programme address? I agree to be contacted about this request. Book a scoping call", + "links": [ + { + "text": "Explore Step 2: AI engineer conversion →", + "url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/" + }, + { + "text": "Step 2: AI engineer conversion", + "url": "https://towardsai.com/enterprise/software-developer-to-ai-engineer/" + } + ], + "chunks": [ + { + "chunk_id": "tai-baed956fd92b9d455e105e0f", + "text": "Step 1 of 2 · Every developer\n\nYour team has coding-agent seats. Now make them pay.\nCoding agents amplify the discipline a team already has. We install that discipline in one day, inside your approved Claude Code or Codex setup. Start with one day Book a scoping call 30 minutes with the instructors who deliver it. No sales team. 8x more code merged per engineer at Anthropic since 2024, with trained agent use −19% experienced devs slower with AI when untrained, while believing they were faster (METR) 1 day to install the shared operating practice, on-site or remote, up to 40 developers Programme at a glance Start 1-day bootcamp , up to 40 developers Scale 2–3-day deep dive · per-seat licence Tools Claude Code · Codex, your approved setup Where On-site or remote, English or French Taught by Engineers who ship with these tools daily Then Debrief names your Step-2 AI engineer candidates Training delivered for teams at 01 Start here\n\nPick your starting point\nThree ways in, one operating standard. Most teams prove it in one day, then scale. Every bootcamp ends with a written capability debrief and our recommendation on whether to go further, including “don't.” 1 day Start here\n\nOne-day agentic developer bootcamp\nOne concentrated day inside your approved setup. The operating loop of plan, direct, review, and verify, plus repository instructions (CLAUDE.md / AGENTS.md) drafted against one of your real repos, and your team conventions v1. Cohort Up to 40 developers Leaves with Working setups, drafted agent files, conventions v1, debrief Best for Proving trained agent use to a sceptical team, fast Fixed fee per cohort · quoted at scoping Book the day → 2–3 days Most chosen", + "heading": "Your team has coding-agent seats. Now make them pay.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-12326793a701626724b20342", + "text": "Step 1 of 2 · Every developer" + }, + { + "span_id": "span-4af2bd237fd6448a41b08f6e", + "text": "Coding agents amplify the discipline a team already has." + }, + { + "span_id": "span-494466dcaee3c4874c9cad65", + "text": "We install that discipline in one day, inside your approved Claude Code or Codex setup." + }, + { + "span_id": "span-65b971a60d2c4675b8ba4f2d", + "text": "Start with one day Book a scoping call" + }, + { + "span_id": "span-a74a1e2ea33ae0facbe510be", + "text": "30 minutes with the instructors who deliver it." + }, + { + "span_id": "span-99fd85c328e70a134f9d4eda", + "text": "No sales team." + }, + { + "span_id": "span-c44e1464d26bbd81a4aaf97b", + "text": "8x more code merged per engineer at Anthropic since 2024, with trained agent use" + }, + { + "span_id": "span-d45c96e3607327cf8b9c9109", + "text": "−19% experienced devs slower with AI when untrained, while believing they were faster (METR)" + }, + { + "span_id": "span-dd11477a46133fb5d7b2fc6d", + "text": "1 day to install the shared operating practice, on-site or remote, up to 40 developers" + }, + { + "span_id": "span-8434fe92c192d265a4c007b1", + "text": "Programme at a glance" + }, + { + "span_id": "span-c234b6a1bd33a2608d1bc399", + "text": "1-day bootcamp , up to 40 developers" + }, + { + "span_id": "span-a49afef1edb71df4e5689a80", + "text": "2–3-day deep dive · per-seat licence" + }, + { + "span_id": "span-42a4d5cecf92034cf6e63d36", + "text": "Claude Code · Codex, your approved setup" + }, + { + "span_id": "span-4e518b8d248d2fe83a57c633", + "text": "On-site or remote, English or French" + }, + { + "span_id": "span-34b17abaf13a413aa0a5660f", + "text": "Taught by" + }, + { + "span_id": "span-dfab95409ca6c0c3c85aa26a", + "text": "Engineers who ship with these tools daily" + }, + { + "span_id": "span-75202a6c8b409c0edd3aef85", + "text": "Debrief names your Step-2 AI engineer candidates" + }, + { + "span_id": "span-dc33e7bf76a47911ca155ee0", + "text": "Training delivered for teams at" + }, + { + "span_id": "span-7c3b36f32c983c2841a85b2e", + "text": "01 Start here" + }, + { + "span_id": "span-243be860d96ae6a33c3842ca", + "text": "Three ways in, one operating standard." + }, + { + "span_id": "span-98d36a1de61545fd93f68890", + "text": "Most teams prove it in one day, then scale." + }, + { + "span_id": "span-2bd4911d4bfa4f2207098c5a", + "text": "Every bootcamp ends with a written capability debrief and our recommendation on whether to go further, including “don't.”" + }, + { + "span_id": "span-d5f1ee88dd6ae0f76450e65e", + "text": "1 day Start here" + }, + { + "span_id": "span-9a234120d9b456f4fb9de65d", + "text": "One concentrated day inside your approved setup." + }, + { + "span_id": "span-48b7dfe78e6fc123dd78e98a", + "text": "The operating loop of plan, direct, review, and verify, plus repository instructions (CLAUDE.md / AGENTS.md) drafted against one of your real repos, and your team conventions v1." + }, + { + "span_id": "span-c6f53b8503d7273618836fea", + "text": "Cohort Up to 40 developers Leaves with Working setups, drafted agent files, conventions v1, debrief Best for Proving trained agent use to a sceptical team, fast" + }, + { + "span_id": "span-d06d294251cdc0df955ad22f", + "text": "Fixed fee per cohort · quoted at scoping Book the day →" + }, + { + "span_id": "span-6f8f724a8123ec40c0ff1f49", + "text": "2–3 days Most chosen" + } + ] + }, + { + "chunk_id": "tai-dc63d598f1dd15cde26f83a0", + "text": "Deep-dive team bootcamp\nTwo instructor-led days across all eight modules: agent files, plan mode, large codebases, multi-agent and MCP, guardrails, cost. Optional applied day on tasks from your own roadmap. Cohort 10 to 40 engineers; repeatable across teams Leaves with Team standards, review discipline, multi-agent practice, debrief Best for Teams defining shared conventions properly Per cohort · quoted after scoping Scope the bootcamp → Per seat\n\nEnterprise Academy licence\nThe full self-paced Academy bundle, including the Agentic Development course, with role-based paths, live kickoff, monthly mentorship clinics, office hours, and leadership analytics. The baseline for hundreds to thousands of developers. Scale Hundreds to thousands of seats Leaves with One shared foundation, progress analytics, lifetime updates Best for Consistent baseline across a large org Per seat · volume bands Discuss seats → Also available: 1:1 coaching for the lead writing your conventions. Cost is driven by participant count, live days, and customisation depth; an indicative quote accompanies the scoping agenda, before you commit to anything. 02 The system", + "heading": "Enterprise Academy licence", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-eb03b044bf70851a08bee978", + "text": "Two instructor-led days across all eight modules: agent files, plan mode, large codebases, multi-agent and MCP, guardrails, cost." + }, + { + "span_id": "span-e25d2dd5647d864ebff13ee2", + "text": "Optional applied day on tasks from your own roadmap." + }, + { + "span_id": "span-09cf05d2574f6dc6367a3193", + "text": "Cohort 10 to 40 engineers; repeatable across teams Leaves with Team standards, review discipline, multi-agent practice, debrief Best for Teams defining shared conventions properly" + }, + { + "span_id": "span-2c17f5811f696f3e4d8371a1", + "text": "Per cohort · quoted after scoping Scope the bootcamp →" + }, + { + "span_id": "span-5b2a159601195a3c89abc2be", + "text": "Per seat" + }, + { + "span_id": "span-1c1fecec224b1864352b13f9", + "text": "The full self-paced Academy bundle, including the Agentic Development course, with role-based paths, live kickoff, monthly mentorship clinics, office hours, and leadership analytics." + }, + { + "span_id": "span-caf93ce1df607c8547cadfc0", + "text": "The baseline for hundreds to thousands of developers." + }, + { + "span_id": "span-0b1376434fb44e224d251907", + "text": "Scale Hundreds to thousands of seats Leaves with One shared foundation, progress analytics, lifetime updates Best for Consistent baseline across a large org" + }, + { + "span_id": "span-16a950052a298bdd0d197202", + "text": "Per seat · volume bands Discuss seats →" + }, + { + "span_id": "span-e0ad1f1d59141fbb6df4c333", + "text": "Also available: 1:1 coaching for the lead writing your conventions." + }, + { + "span_id": "span-6be3d923dfa75e6c2f490a80", + "text": "Cost is driven by participant count, live days, and customisation depth; an indicative quote accompanies the scoping agenda, before you commit to anything." + }, + { + "span_id": "span-9b38b06ca644c0bd86951016", + "text": "02 The system" + } + ] + }, + { + "chunk_id": "tai-f55baa08499a607fc1f2ec1d", + "text": "Two steps: use AI to build software, then build AI into it\nStep 1 makes every developer effective with coding agents and reveals who should go further. Step 2 converts your strongest builders into AI engineers. You are on the Step 1 page. Step 1 · Agentic developers ← this page Step 2 · AI engineers Who Every software developer , no AI background needed Selected builders, typically the top 10–20% They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems Outcome Ship faster on today's roadmap ; agent code passes the same bar as human code New products and AI systems; the capability you cannot hire Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · senior residency · per-seat licence When Now: it pays back on work already in flight After Step 1: the debrief names who's ready Explore Step 2: AI engineer conversion → Same instructors, same standard, one scoping call covers both. 03 Why it works", + "heading": "Two steps: use AI to build software, then build AI into it", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-1ab8427f649ef17a08d76abf", + "text": "Step 1 makes every developer effective with coding agents and reveals who should go further." + }, + { + "span_id": "span-641a920df3cbf9c3d346be2e", + "text": "Step 2 converts your strongest builders into AI engineers." + }, + { + "span_id": "span-5e53498cb42dd7dd80aef6ff", + "text": "You are on the Step 1 page." + }, + { + "span_id": "span-95f331daab17260788118b1e", + "text": "Who Every software developer , no AI background needed Selected builders, typically the top 10–20%" + }, + { + "span_id": "span-4a74d492e79b8920382a6150", + "text": "They learn To plan, direct, review, and verify coding-agent work under one standard To design, evaluate, ship, and maintain production LLM systems" + }, + { + "span_id": "span-7f0359fdbd21e4d014940be2", + "text": "Outcome Ship faster on today's roadmap ; agent code passes the same bar as human code New products and AI systems; the capability you cannot hire" + }, + { + "span_id": "span-e7275d845348c2a28127a716", + "text": "Format 1-day or 2–3-day bootcamp · per-seat licence Part-time cohort waves · senior residency · per-seat licence" + }, + { + "span_id": "span-444eb04c0a3ceabc02882809", + "text": "When Now: it pays back on work already in flight After Step 1: the debrief names who's ready" + }, + { + "span_id": "span-0f8f794176e3cc0c6b348cd0", + "text": "Explore Step 2: AI engineer conversion → Same instructors, same standard, one scoping call covers both." + }, + { + "span_id": "span-a6f9b9bcae05154103ac1ca8", + "text": "03 Why it works" + } + ] + }, + { + "chunk_id": "tai-74baf76a4a2551652a4407ea", + "text": "Seats are everywhere. Skill is not.\nYour developers already spend a median of two hours a day with AI. The field data is blunt about what separates the teams that win from the teams that stall. AI is an amplifier DORA's 2025 study of ~5,000 professionals: 90% of developers now use AI, a median of two hours a day, and over 80% report productivity gains, yet delivery instability rises unless the workflow is redesigned. Seven organisational capabilities decide whether individual gains ever reach the org. Amplifiers make disciplined teams faster and undisciplined teams slower. DORA 2025 · Google Cloud Untrained use backfires METR found experienced developers 19% slower with AI on familiar code, while believing they were 20% faster. Self-assessment is miscalibrated, so self-learning doesn't converge. 66% of developers fight “almost right” output; 45% say debugging AI code takes longer; only around one in six report significant productivity gains. METR 2025 · Stack Overflow 2025 The gap is trainable Inside Anthropic, Claude now authors 80%+ of merged code and code merged per engineer is up 8x since 2024. Experts drive ~12 agent actions and 3,200 words of output per prompt versus ~5 and 600 for novices. The difference is task framing, context, and verification: exactly what we train. BCG finds 5+ hours of structured training with coaching is what turns tool access into regular, effective use. Anthropic 2026 · BCG 2025 Waiting compounds the gap Google's CEO says AI now writes around three-quarters of the company's new code. Microsoft has told staff AI use is “no longer optional.” Mandates without enablement backfire; the winners build capability instead of compulsion, and the first trained cohort sets the standard every later cohort inherits. Alphabet 2026 · Microsoft 2025 04", + "heading": "Seats are everywhere. Skill is not.", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-50244f65bf04c4b80b6f76fa", + "text": "Your developers already spend a median of two hours a day with AI." + }, + { + "span_id": "span-21411866a5be083e7de4ec70", + "text": "The field data is blunt about what separates the teams that win from the teams that stall." + }, + { + "span_id": "span-db4dd8527605a1fa241e8713", + "text": "AI is an amplifier DORA's 2025 study of ~5,000 professionals: 90% of developers now use AI, a median of two hours a day, and over 80% report productivity gains, yet delivery instability rises unless the workflow is redesigned." + }, + { + "span_id": "span-62efebacea09eaddd3ae6a5d", + "text": "Seven organisational capabilities decide whether individual gains ever reach the org." + }, + { + "span_id": "span-da2f57c68b16b88b42c2b396", + "text": "Amplifiers make disciplined teams faster and undisciplined teams slower." + }, + { + "span_id": "span-f99a15d2a2bce5affb0916ca", + "text": "DORA 2025 · Google Cloud" + }, + { + "span_id": "span-2d65023350d19f18aa6f58ca", + "text": "Untrained use backfires METR found experienced developers 19% slower with AI on familiar code, while believing they were 20% faster." + }, + { + "span_id": "span-4d4fd55d664e298d13017f32", + "text": "Self-assessment is miscalibrated, so self-learning doesn't converge." + }, + { + "span_id": "span-612312a24751ed577c2ee304", + "text": "66% of developers fight “almost right” output; 45% say debugging AI code takes longer; only around one in six report significant productivity gains." + }, + { + "span_id": "span-15e60a993d95209d92f93d95", + "text": "METR 2025 · Stack Overflow 2025" + }, + { + "span_id": "span-714265dbe0fa0b6e29272854", + "text": "The gap is trainable Inside Anthropic, Claude now authors 80%+ of merged code and code merged per engineer is up 8x since 2024." + }, + { + "span_id": "span-4c8715b1dae7f047ead56f49", + "text": "Experts drive ~12 agent actions and 3,200 words of output per prompt versus ~5 and 600 for novices." + }, + { + "span_id": "span-18034355bda09d2d2cc9990c", + "text": "The difference is task framing, context, and verification: exactly what we train." + }, + { + "span_id": "span-f64a16b6a887e4bfc278232d", + "text": "BCG finds 5+ hours of structured training with coaching is what turns tool access into regular, effective use." + }, + { + "span_id": "span-bf88011e840c54732135ee3d", + "text": "Anthropic 2026 · BCG 2025" + }, + { + "span_id": "span-a43e35664f4e2b75a599bd00", + "text": "Waiting compounds the gap Google's CEO says AI now writes around three-quarters of the company's new code." + }, + { + "span_id": "span-af93950d617f45e2a83b30f4", + "text": "Microsoft has told staff AI use is “no longer optional.” Mandates without enablement backfire; the winners build capability instead of compulsion, and the first trained cohort sets the standard every later cohort inherits." + }, + { + "span_id": "span-d98a15cbecee5714ac9ce6be", + "text": "Alphabet 2026 · Microsoft 2025" + } + ] + }, + { + "chunk_id": "tai-52f85c12b2058bd38389d89c", + "text": "Seats are everywhere. Skill is not.\nStraight answers", + "heading": "Seats are everywhere. Skill is not.", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-c924ee79294d46168ae80c8e", + "text": "Seats are everywhere." + }, + { + "span_id": "span-cf547c79d5fff5efccf85236", + "text": "Skill is not." + } + ] + }, + { + "chunk_id": "tai-32887e67e2ec2e74374dab49", + "text": "The questions your CFO will ask\n“Our devs can self-learn from docs and YouTube.” They've had three years. Only ~16% of developers report significant productivity gains from AI, and METR shows devs believed they were 20% faster while measuring 19% slower: miscalibrated self-assessment means self-learning doesn't converge . Structured training with coaching is what moves usage (BCG 2025). “The tools change too fast; training goes stale.” The durable skills are tool-agnostic: planning, context engineering, review discipline, verification, team conventions. They transfer across Claude Code, Codex, and whatever ships next. We re-train on the frontier monthly because we ship with it , and Academy content carries lifetime updates. “We can't take engineers off delivery.” The entry format is one day . Your engineers already spend a median two hours a day with AI, unmanaged. The bootcamp redirects time already being spent, and applied exercises run on tasks from your own roadmap, not toy repos. “Security won't allow it.” Everything runs in your approved environment : your tool surface, your permissions, no production access required, NDAs standard. We've delivered inside Europol. And the biggest security risk is untrained use; review boundaries and guardrails are half the curriculum. “We already pay for seats, so we're covered.” Seats are access, not capability. DORA's finding is that AI amplifies the engineering discipline you already have ; licences amplify nothing on their own. The 8x gains show up where trained use meets shared standards. “How do we know it worked?” We capture a baseline in scoping and close with a written capability debrief : what changed in how your engineers work, the conventions your team keeps, and the metrics to watch: throughput, review load, spend per", + "heading": "The questions your CFO will ask", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-0d40826795314ddae4c37c0c", + "text": "“Our devs can self-learn from docs and YouTube.” They've had three years." + }, + { + "span_id": "span-3fd2f92b12ce4f42cdd804a2", + "text": "Only ~16% of developers report significant productivity gains from AI, and METR shows devs believed they were 20% faster while measuring 19% slower: miscalibrated self-assessment means self-learning doesn't converge ." + }, + { + "span_id": "span-93b1ba657453b89a884d34b7", + "text": "Structured training with coaching is what moves usage (BCG 2025)." + }, + { + "span_id": "span-7d99ac311dfc34eec4e9a37d", + "text": "“The tools change too fast; training goes stale.” The durable skills are tool-agnostic: planning, context engineering, review discipline, verification, team conventions." + }, + { + "span_id": "span-38dfb95f443218a1bd15f068", + "text": "They transfer across Claude Code, Codex, and whatever ships next." + }, + { + "span_id": "span-6f08ff761a3d54eb50617303", + "text": "We re-train on the frontier monthly because we ship with it , and Academy content carries lifetime updates." + }, + { + "span_id": "span-70a5e622a4920aeb281f1a48", + "text": "“We can't take engineers off delivery.” The entry format is one day ." + }, + { + "span_id": "span-e7b03d0107632f1b243b9d19", + "text": "Your engineers already spend a median two hours a day with AI, unmanaged." + }, + { + "span_id": "span-55ff247eab7f6689d34506e6", + "text": "The bootcamp redirects time already being spent, and applied exercises run on tasks from your own roadmap, not toy repos." + }, + { + "span_id": "span-602291620d6380f39a7a7f42", + "text": "“Security won't allow it.” Everything runs in your approved environment : your tool surface, your permissions, no production access required, NDAs standard." + }, + { + "span_id": "span-9466204f0ebf1c05bc57b61d", + "text": "We've delivered inside Europol." + }, + { + "span_id": "span-9283c639bb7d565106c0ff8a", + "text": "And the biggest security risk is untrained use; review boundaries and guardrails are half the curriculum." + }, + { + "span_id": "span-f2f156d2d8a700c862510f46", + "text": "“We already pay for seats, so we're covered.” Seats are access, not capability." + }, + { + "span_id": "span-7721b70a3e77e03ef702a442", + "text": "DORA's finding is that AI amplifies the engineering discipline you already have ; licences amplify nothing on their own." + }, + { + "span_id": "span-dff494471110f42aeeaaf2f4", + "text": "The 8x gains show up where trained use meets shared standards." + } + ] + }, + { + "chunk_id": "tai-efae832f018f1b2011945887", + "text": "The questions your CFO will ask\nmerged change. Scale to more cohorts only if the debrief earns it. 05 Delivered in the field\n\nTrained where mistakes aren't an option\nEuropol Law enforcement · EU Three-day security-first bootcamp: technical investigation teams went from cautious LLM users to confident builders of secure, reviewable intelligence pipelines. Outcome: special agents deploying AI across operational intelligence workflows. Major US public library system Public sector · USA Four-day, two-track programme for engineers and operations staff, backed by 43 hands-on notebooks and a hackathon on real pain points. Outcome: adoption from foundations to shipped build, owned by the library's own teams, not a vendor. Global banks & leading funds Finance Trader training at a leading global bank; live Claude Cowork sessions for investment teams at leading funds; enterprise bootcamps tailored to each firm's stack. Outcome: desk-level AI workflows in regulated, latency-sensitive environments. “Since the training, I’ve built two local apps using your vibe-coding approach. With every Gemini release I test vibe coding capabilities and have been unimpressed. Your approach is brilliant… We’re going to make it our primary developer model.” VP of IT Major US public library system “Excellent in-depth handling of trade-offs in evaluating and deploying agent-based solutions. I've used a lot of the advice in building our enterprise practices to agent use in my organisation.” Cathal Curtin Student, Agent Engineering 06 Who delivers", + "heading": "Trained where mistakes aren't an option", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-bb6977e3b16452bf03f8ca6b", + "text": "Scale to more cohorts only if the debrief earns it." + }, + { + "span_id": "span-783cfc0ebb370a52d2fc10cb", + "text": "05 Delivered in the field" + }, + { + "span_id": "span-8c8160b0e9118a463f390583", + "text": "Europol Law enforcement · EU" + }, + { + "span_id": "span-e76b249d9799a7d86f69f664", + "text": "Three-day security-first bootcamp: technical investigation teams went from cautious LLM users to confident builders of secure, reviewable intelligence pipelines." + }, + { + "span_id": "span-8d38c10bae9c892facaeeac7", + "text": "Outcome: special agents deploying AI across operational intelligence workflows." + }, + { + "span_id": "span-036c43359acb12aac2b78211", + "text": "Major US public library system Public sector · USA" + }, + { + "span_id": "span-93378756884240631ff23fc0", + "text": "Four-day, two-track programme for engineers and operations staff, backed by 43 hands-on notebooks and a hackathon on real pain points." + }, + { + "span_id": "span-219a61a744700556bc9d45cf", + "text": "Outcome: adoption from foundations to shipped build, owned by the library's own teams, not a vendor." + }, + { + "span_id": "span-ec7f8a1cf9317b078fce0809", + "text": "Global banks & leading funds Finance" + }, + { + "span_id": "span-a6ed6afd9e8bdec2cf942515", + "text": "Trader training at a leading global bank; live Claude Cowork sessions for investment teams at leading funds; enterprise bootcamps tailored to each firm's stack." + }, + { + "span_id": "span-2053ad0ad3a02c14b4481dce", + "text": "Outcome: desk-level AI workflows in regulated, latency-sensitive environments." + }, + { + "span_id": "span-426e5ffec068170c12368255", + "text": "“Since the training, I’ve built two local apps using your vibe-coding approach." + }, + { + "span_id": "span-666d053c4cc97e4754043ecd", + "text": "With every Gemini release I test vibe coding capabilities and have been unimpressed." + }, + { + "span_id": "span-ee7be3af62682ed6a70ccc64", + "text": "Your approach is brilliant… We’re going to make it our primary developer model.”" + }, + { + "span_id": "span-a61122df123276b5b6293a2c", + "text": "VP of IT" + }, + { + "span_id": "span-bb6274e87ac6c224e44b2b88", + "text": "Major US public library system" + }, + { + "span_id": "span-ace66b94d8d9c8d7959861a6", + "text": "“Excellent in-depth handling of trade-offs in evaluating and deploying agent-based solutions." + }, + { + "span_id": "span-74e3afdeba9d9ca784f6ac0d", + "text": "I've used a lot of the advice in building our enterprise practices to agent use in my organisation.”" + }, + { + "span_id": "span-afd31151412a324f7cc51b03", + "text": "Cathal Curtin" + }, + { + "span_id": "span-9ac9f6c3c910a094f454fd68", + "text": "Student, Agent Engineering" + }, + { + "span_id": "span-a278956c5eb6407af5225636", + "text": "06 Who delivers" + } + ] + }, + { + "chunk_id": "tai-bfb45843df2b1aff8f0b6d84", + "text": "No training faculty. Practising engineers.\nThe people in the room ship client systems with coding agents every working week. The material is the conventions, failure modes, and review habits that survived real codebases. LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production. Newsletter read by 200k+ practitioners. Ex-J.P. Morgan VP. LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA, Yoshua Bengio's lab. FC Fabio Chiusano Senior AI Engineer NLP lead; ships the client systems the curriculum is drawn from. SV Samridhi Vaid Machine Learning Engineer Ships production multimodal AI, RAG, and agent systems; M.Sc. in Computing Science, University of Alberta. 500k+ AI practitioners taught since 2019 15+ in-house AI engineers shipping client systems 10k+ copies of Building LLMs for Production sold 21+ AI applications shipped to production 07 How it runs\n\nLow commitment in, evidence out\n01\n\nScoping call · 30 min\nTeams, approved tools, working patterns, and priorities, discussed with the instructors. You get a written plan and an indicative quote within one business day. 02\n\nBaseline\nWe capture where the team actually is (usage patterns, conventions, metrics worth watching) so the debrief measures change, not vibes. 03\n\nThe day(s)\nLive workshops inside your approved environment. No surprise tooling, no unapproved repository access, acceptance and merge stay with your engineers. 04", + "heading": "No training faculty. Practising engineers.", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-ab5d642b0c8fce5457d74e18", + "text": "The people in the room ship client systems with coding agents every working week." + }, + { + "span_id": "span-9d85afa96b962a1e1889ef0d", + "text": "The material is the conventions, failure modes, and review habits that survived real codebases." + }, + { + "span_id": "span-aac12878c5bc642130d323c7", + "text": "LP Louie Peters Co-CEO · Instructor Co-author, Building LLMs for Production." + }, + { + "span_id": "span-ba4e8a50bb2a52d805a46d37", + "text": "Newsletter read by 200k+ practitioners." + }, + { + "span_id": "span-16b367e42a95f6a37aefb13a", + "text": "Ex-J.P." + }, + { + "span_id": "span-904d44b057b65902d1190157", + "text": "Morgan VP." + }, + { + "span_id": "span-29357c9fcc92ef2da2a59b7a", + "text": "LB Louis-François Bouchard CTO · Instructor Co-author of the book; 210k followers; ex-PhD at MILA, Yoshua Bengio's lab." + }, + { + "span_id": "span-a37f023297238a25d15a1c78", + "text": "FC Fabio Chiusano Senior AI Engineer NLP lead; ships the client systems the curriculum is drawn from." + }, + { + "span_id": "span-b7a108453a2352925757db4f", + "text": "SV Samridhi Vaid Machine Learning Engineer Ships production multimodal AI, RAG, and agent systems; M.Sc. in Computing Science, University of Alberta." + }, + { + "span_id": "span-ec932e0db7d167603bd114ce", + "text": "500k+ AI practitioners taught since 2019" + }, + { + "span_id": "span-096b63be816921b7e58d4984", + "text": "15+ in-house AI engineers shipping client systems" + }, + { + "span_id": "span-399c0f4094b18c6a6423d8f7", + "text": "10k+ copies of Building LLMs for Production sold" + }, + { + "span_id": "span-d4bb8cc4f4965708d0a5cd1b", + "text": "21+ AI applications shipped to production" + }, + { + "span_id": "span-57fdb58f73df8c07d5d348ea", + "text": "07 How it runs" + }, + { + "span_id": "span-24fb3e3139d4fb9958436384", + "text": "Low commitment in, evidence out" + }, + { + "span_id": "span-3d845c7d3226104e311ffb41", + "text": "Scoping call · 30 min" + }, + { + "span_id": "span-7023f677590d46edd11a04dd", + "text": "Teams, approved tools, working patterns, and priorities, discussed with the instructors." + }, + { + "span_id": "span-09306fcf13198d9c97c6d178", + "text": "You get a written plan and an indicative quote within one business day." + }, + { + "span_id": "span-1b59ee5592bad927d0f60fb1", + "text": "We capture where the team actually is (usage patterns, conventions, metrics worth watching) so the debrief measures change, not vibes." + }, + { + "span_id": "span-85bb7c22ce4f5db56707fecb", + "text": "Live workshops inside your approved environment." + }, + { + "span_id": "span-f3906b46e392ac09313d7ce4", + "text": "No surprise tooling, no unapproved repository access, acceptance and merge stay with your engineers." + } + ] + }, + { + "chunk_id": "tai-6ad3bb671a2dddf7d2ed29f1", + "text": "Debrief & decision\nA written capability debrief: what changed, which conventions to keep, who's ready for Step 2, and whether further cohorts are worth it. Scale only if it earns it. + Start with one day. Prove it on one cohort before you commit to a programme. + Your tools, your rules. Scoped to the tool surface your organisation has approved. No production access required. + Procurement-ready. NDAs on request, PO and invoicing supported, references in scoping. + We'll say if it's not for you. If a cheaper format fits, or training isn't the right tool, the instructors tell you on the call.\n\nOne day. Up to 40 developers. Your approved tools.\nBook the scoping call; get the plan and quote within one business day. Book a scoping call 08 Logistics", + "heading": "Debrief & decision", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-b85059309b240fc1e72d35d2", + "text": "A written capability debrief: what changed, which conventions to keep, who's ready for Step 2, and whether further cohorts are worth it." + }, + { + "span_id": "span-99c08abc539e5266d8652ab0", + "text": "Scale only if it earns it." + }, + { + "span_id": "span-3de18cedfede354de5155f28", + "text": "+ Start with one day." + }, + { + "span_id": "span-9a87f94968f923fd47578f41", + "text": "Prove it on one cohort before you commit to a programme." + }, + { + "span_id": "span-68d738b7ec5ee61680bc619d", + "text": "+ Your tools, your rules." + }, + { + "span_id": "span-d2dd352a5d6ef0e9f0911cfd", + "text": "Scoped to the tool surface your organisation has approved." + }, + { + "span_id": "span-29c1d6ec920dc50a745dd03a", + "text": "No production access required." + }, + { + "span_id": "span-86313f0e1aae8d66ac1e7a30", + "text": "+ Procurement-ready." + }, + { + "span_id": "span-e61b9aef4637fbc18c1a387a", + "text": "NDAs on request, PO and invoicing supported, references in scoping." + }, + { + "span_id": "span-f7d10c7bc54077a941172e23", + "text": "+ We'll say if it's not for you." + }, + { + "span_id": "span-a33d4d9f8a2ef88c75c26a59", + "text": "If a cheaper format fits, or training isn't the right tool, the instructors tell you on the call." + }, + { + "span_id": "span-c49c2c03af0d7d59b2921694", + "text": "Up to 40 developers." + }, + { + "span_id": "span-1c8658e39d050d9f568bea57", + "text": "Book the scoping call; get the plan and quote within one business day." + }, + { + "span_id": "span-39bcad7546057007275b67b7", + "text": "Book a scoping call" + }, + { + "span_id": "span-8911af0d80aa7c8f40d8b8fc", + "text": "08 Logistics" + } + ] + }, + { + "chunk_id": "tai-079eee58acf6754cd8d41095", + "text": "Frequently asked questions\nDoes the programme require one specific coding agent? + No. We scope workshops around the approved tool surface relevant to your team (Claude Code, Codex, or both) and explain differences where practices are tool-specific, including how to standardise across them. What exactly happens on the applied day? + Only with agreement: after sign-off on approved repositories or prepared tasks, permissions, and review expectations, the cohort works agreed tasks from your roadmap under instructor supervision. It's applied learning, never production delivery. One signature session: an hour where engineers may only plan, direct, and review while the agent writes everything; the diffs are then reviewed against your own bar. What are the prerequisites? + Software engineering experience and access to your approved coding-agent tooling. No AI background needed; that's the point of Step 1. What comes after? + The debrief names next steps: further cohorts, team conventions to keep, and the developers ready for Step 2: AI engineer conversion . Beyond training, the same engineers build custom AI systems and provide fractional AI leadership. 09 Next step", + "heading": "Frequently asked questions", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-66e6ed902531f98c4a013a49", + "text": "We scope workshops around the approved tool surface relevant to your team (Claude Code, Codex, or both) and explain differences where practices are tool-specific, including how to standardise across them." + }, + { + "span_id": "span-d4ad6dbded7a5635bb54a71f", + "text": "Software engineering experience and access to your approved coding-agent tooling." + }, + { + "span_id": "span-c7272863aca9c5a7c3561c4a", + "text": "No AI background needed; that's the point of Step 1." + }, + { + "span_id": "span-2f6edee9628f99d8c6c3bb5d", + "text": "The debrief names next steps: further cohorts, team conventions to keep, and the developers ready for Step 2: AI engineer conversion ." + }, + { + "span_id": "span-8427d03ebb535d7e186b5af4", + "text": "Beyond training, the same engineers build custom AI systems and provide fractional AI leadership." + }, + { + "span_id": "span-13ac5512596180247e62debd", + "text": "09 Next step" + } + ] + }, + { + "chunk_id": "tai-e9d683d4915f1b5439d1e1d8", + "text": "Book a scoping call\nTell us about your developer cohorts, approved tools, and priorities. We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the workshops. 1 On the call: teams, stacks, baseline, and which format fits. 30 minutes, with instructors. 2 You leave with: a written plan, an indicative quote, and a measurement approach. 3 No pressure: if one day is enough, we say so. Every enquiry is read by the Towards AI team. Prefer email? louis@towardsai.net Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet One-day agentic bootcamp Deep-dive bootcamp (2-3 days) Enterprise Academy licence Leader coaching Step 2: AI engineer programme What should the programme address? I agree to be contacted about this request. Book a scoping call", + "heading": "Book a scoping call", + "index": 10, + "evidence_spans": [ + { + "span_id": "span-a11232310badb2a68d1f6e17", + "text": "Tell us about your developer cohorts, approved tools, and priorities." + }, + { + "span_id": "span-8d5259ab09a7c631f06d38ad", + "text": "We reply within one business day with a short scoping agenda and an indicative quote, and you'll talk with the instructors who deliver the workshops." + }, + { + "span_id": "span-eccf96ff7c34e93d77ccfbcb", + "text": "1 On the call: teams, stacks, baseline, and which format fits." + }, + { + "span_id": "span-487b15fa7e2949fec26e83f5", + "text": "30 minutes, with instructors." + }, + { + "span_id": "span-40c305a795edb4077d57f49e", + "text": "2 You leave with: a written plan, an indicative quote, and a measurement approach." + }, + { + "span_id": "span-91be3322a4d7324fdc1dec0e", + "text": "3 No pressure: if one day is enough, we say so." + }, + { + "span_id": "span-4c913c4dff82b596feb6b59d", + "text": "Every enquiry is read by the Towards AI team." + }, + { + "span_id": "span-b94b6bbaea2aa3b5f76c5ceb", + "text": "Prefer email? louis@towardsai.net" + }, + { + "span_id": "span-3fbe4e2e281d4769cd9f7810", + "text": "Full name Work email Company Role Engineering team size Select one 1-10 11-50 51-250 251-1000 1000+ Format of interest Not sure yet One-day agentic bootcamp Deep-dive bootcamp (2-3 days) Enterprise Academy licence Leader coaching Step 2: AI engineer programme What should the programme address?" + }, + { + "span_id": "span-bdd6de9b7f7cae62726705c4", + "text": "I agree to be contacted about this request." + }, + { + "span_id": "span-833f709cc04b8b14b3da3ebb", + "text": "Book a scoping call" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "f4d655bc23f2621d7d54ebbefe2912f22b465c8e177a9c798bbe3c5be412739f", + "content_hash": "f4d655bc23f2621d7d54ebbefe2912f22b465c8e177a9c798bbe3c5be412739f", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/enterprise/agentic-developer-conversion/" + ], + "evidence_hash": "974158e161dec3e8c32df82a969e79d54283179f266554b4edb160f5ff52cb5e" + }, + { + "discovered_url": "https://towardsai.com/academy/affiliate/", + "url": "https://towardsai.com/academy/affiliate/", + "canonical_url": "https://towardsai.com/academy/affiliate/", + "host": "towardsai.com", + "path": "/academy/affiliate", + "kind": "page", + "offer_id": null, + "entity_id": "page:towardsai.com/academy/affiliate", + "title": "Referral & Affiliate Program · Towards AI Academy", + "review_title": "Affiliate", + "reviewed_summary": "Towards AI referral and affiliate program.", + "meta_description": "Share Towards AI Academy courses and earn cashback, free courses, and 20% affiliate revenue. Tiered rewards for casual referrers and creators with an audience.", + "headings": [ + "Become a Towards AI Academy partner", + "Referral reward tiers", + "Why become a referral partner", + "Your rewards are courses you can actually use", + "Help your circle gain ground", + "Be recognized as the one who saw it early", + "Earn ongoing perks every time you share", + "Promotion made easy", + "Marketing assets provided", + "Ready-to-send emails", + "Course details upfront", + "Conversion-optimized materials", + "Already have an audience?", + "Frequently asked questions", + "Start earning with the courses you already trust" + ], + "text": "Referral & affiliate program\n\nBecome a Towards AI Academy partner\nShare our courses and unlock advanced training, cashback, and affiliate revenue. Start referring now Join affiliate (creators only) 20% revenue share for Gold affiliates $180+ commission per bundle sale 15% cashback from your first referral 01 How rewards scale\n\nReferral reward tiers\nBronze Silver Gold Requirement 1 course referred 3 courses or 1 bundle 10 courses or 5 bundles Your reward 15% cashback on your own course (up to $60 per course) Unlock any course of your choice Full access to all current offerings For your referrals 15% discount for your friends 15% discount for your friends Entry into our 20% revenue-share affiliate program * Your total cashback can't exceed what you've paid for your own course(s). 02 The case for sharing\n\nWhy become a referral partner\nTurn sharing into a lasting advantage. Unlock courses with real payoff\n\nYour rewards are courses you can actually use\nEach referral earns you access to programs that build tangible outcomes: projects you can showcase, skills employers look for, and AI literacy that pays off in real work. Bring others into the AI advantage\n\nHelp your circle gain ground\nAI is changing work at every level. By inviting friends, colleagues, or family, you give them the chance to join the group that thrives in this shift, equipped with skills they can put to use right away. Position yourself at the forefront\n\nBe recognized as the one who saw it early\nWhen you share these opportunities, you establish yourself as the trusted voice who connects people to what matters in AI. That reputation compounds, personally and professionally. Unlock your rewards\n\nEarn ongoing perks every time you share\nJoin our referral program and turn your referrals into free courses and 20% revenue share. Join the program We respect your privacy. Unsubscribe at any time. 03 We do the heavy lifting\n\nPromotion made easy\nYou don't need to be a marketing expert to succeed. 01\n\nMarketing assets provided\nFrom banners to social media posts, the creative work is already done for you. 02\n\nReady-to-send emails\nUse pre-written emails to save time and drive clicks. 03\n\nCourse details upfront\nYour first email includes a breakdown of our courses so you can promote what fits your audience best. 04\n\nConversion-optimized materials\nAll assets are tested and designed to help you earn more. Get brand assets 04 For creators\n\nAlready have an audience?\nReply \"Affiliate\" to the welcome email to get enrolled and start earning. 01 20% commission on every sale. Earn up to $180+ per bundle. 02 Free access to the courses you promote. 03 Private Slack for affiliates. 04 Done-for-you marketing assets. 05 Early access to upcoming courses, launches, and exclusive content. 06 Higher commission tiers available for top affiliates. Potential earnings · Gold tier examples ~$70 per flagship course sale ~$180+ per bundle sale, with higher potential for B2B and team training referrals. One sale is real revenue; many affiliates scale with just a few well-placed links. $18,692.99 in 8 months Shared weekly on YouTube, LinkedIn, newsletters, and learning guides. 66k followers across platforms. “Repeat visibility + multi-channel placement.” $6,730.21 from just 3 posts One technical blog mention and two LinkedIn posts, to 69k followers on a single platform. “Minimal effort, solid return.” 05 Questions\n\nFrequently asked questions\nHow do I join the affiliate program? + Use the signup form above, and you'll get an email with all the details to get started right away. Do I need an audience to earn? + Not at all. You can start by referring friends casually and earn free rewards. How are referrals tracked? + For affiliates, we use Thinkific's built-in tracking. For casual referrers, we verify manually by email. How do I get paid as an affiliate? + Payouts are made monthly via PayPal. Can I do both? + Yes. Start with referral rewards, and once you hit 10 referrals, we'll invite you to the 20% commission program.\n\nStart earning with the courses you already trust\nStart referring now Get brand assets", + "links": [ + { + "text": "Get brand assets", + "url": "https://seldonia.notion.site/Affiliate-Promotion-Hub-255f9b6f42708091ade4c91f75578855" + } + ], + "chunks": [ + { + "chunk_id": "tai-f5316406b01991377e33ea06", + "text": "Referral & affiliate program\n\nBecome a Towards AI Academy partner\nShare our courses and unlock advanced training, cashback, and affiliate revenue. Start referring now Join affiliate (creators only) 20% revenue share for Gold affiliates $180+ commission per bundle sale 15% cashback from your first referral 01 How rewards scale\n\nReferral reward tiers\nBronze Silver Gold Requirement 1 course referred 3 courses or 1 bundle 10 courses or 5 bundles Your reward 15% cashback on your own course (up to $60 per course) Unlock any course of your choice Full access to all current offerings For your referrals 15% discount for your friends 15% discount for your friends Entry into our 20% revenue-share affiliate program * Your total cashback can't exceed what you've paid for your own course(s). 02 The case for sharing\n\nWhy become a referral partner\nTurn sharing into a lasting advantage. Unlock courses with real payoff\n\nYour rewards are courses you can actually use\nEach referral earns you access to programs that build tangible outcomes: projects you can showcase, skills employers look for, and AI literacy that pays off in real work. Bring others into the AI advantage\n\nHelp your circle gain ground\nAI is changing work at every level. By inviting friends, colleagues, or family, you give them the chance to join the group that thrives in this shift, equipped with skills they can put to use right away. Position yourself at the forefront\n\nBe recognized as the one who saw it early\nWhen you share these opportunities, you establish yourself as the trusted voice who connects people to what matters in AI. That reputation compounds, personally and professionally. Unlock your rewards", + "heading": "Referral reward tiers", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-6ca962aaa7508b6c3de31955", + "text": "Referral & affiliate program" + }, + { + "span_id": "span-0583bf5eb6a970ab5f5df752", + "text": "Share our courses and unlock advanced training, cashback, and affiliate revenue." + }, + { + "span_id": "span-c85e649b758e4e50ad8d8519", + "text": "Start referring now Join affiliate (creators only)" + }, + { + "span_id": "span-c01c162aae9605499b6dd9ed", + "text": "20% revenue share for Gold affiliates" + }, + { + "span_id": "span-9ab05213e442c9b1ce1177fc", + "text": "$180+ commission per bundle sale" + }, + { + "span_id": "span-105222c1e7b4cc90a06c14f8", + "text": "15% cashback from your first referral" + }, + { + "span_id": "span-fb969eccd6590e3c03e2ceec", + "text": "01 How rewards scale" + }, + { + "span_id": "span-d6bbc8b2d77e04f740fe5a04", + "text": "Requirement 1 course referred 3 courses or 1 bundle 10 courses or 5 bundles" + }, + { + "span_id": "span-b63363dd6eab8703a5707f37", + "text": "Your reward 15% cashback on your own course (up to $60 per course) Unlock any course of your choice Full access to all current offerings" + }, + { + "span_id": "span-91c176e964cf7683f8ebbd15", + "text": "For your referrals 15% discount for your friends 15% discount for your friends Entry into our 20% revenue-share affiliate program" + }, + { + "span_id": "span-f36a6a4fec5380ecc5a8bb77", + "text": "* Your total cashback can't exceed what you've paid for your own course(s)." + }, + { + "span_id": "span-8e973016dec30cd367343971", + "text": "02 The case for sharing" + }, + { + "span_id": "span-7566985e8c5e2d4deea56ca3", + "text": "Turn sharing into a lasting advantage." + }, + { + "span_id": "span-3d75cc5318a64800a90863f8", + "text": "Unlock courses with real payoff" + }, + { + "span_id": "span-476031a32e95553830195a33", + "text": "Each referral earns you access to programs that build tangible outcomes: projects you can showcase, skills employers look for, and AI literacy that pays off in real work." + }, + { + "span_id": "span-3d0fbe1cd0c6b24243e0510c", + "text": "Bring others into the AI advantage" + }, + { + "span_id": "span-bc6c3e26dd79bd7752b9bbed", + "text": "AI is changing work at every level." + }, + { + "span_id": "span-9868aff013f5dbdba6411393", + "text": "By inviting friends, colleagues, or family, you give them the chance to join the group that thrives in this shift, equipped with skills they can put to use right away." + }, + { + "span_id": "span-11c0d4b190a677e3d878fef3", + "text": "Position yourself at the forefront" + }, + { + "span_id": "span-e2fa18f494e4d366e9226b36", + "text": "When you share these opportunities, you establish yourself as the trusted voice who connects people to what matters in AI." + }, + { + "span_id": "span-28e73a15494eba3bdace4cbb", + "text": "That reputation compounds, personally and professionally." + }, + { + "span_id": "span-08c9d777343b20051b69a0c7", + "text": "Unlock your rewards" + } + ] + }, + { + "chunk_id": "tai-101e6703527eedd968a6bedc", + "text": "Earn ongoing perks every time you share\nJoin our referral program and turn your referrals into free courses and 20% revenue share. Join the program We respect your privacy. Unsubscribe at any time. 03 We do the heavy lifting\n\nPromotion made easy\nYou don't need to be a marketing expert to succeed. 01\n\nMarketing assets provided\nFrom banners to social media posts, the creative work is already done for you. 02\n\nReady-to-send emails\nUse pre-written emails to save time and drive clicks. 03\n\nCourse details upfront\nYour first email includes a breakdown of our courses so you can promote what fits your audience best. 04\n\nConversion-optimized materials\nAll assets are tested and designed to help you earn more. Get brand assets 04 For creators\n\nAlready have an audience?\nReply \"Affiliate\" to the welcome email to get enrolled and start earning. 01 20% commission on every sale. Earn up to $180+ per bundle. 02 Free access to the courses you promote. 03 Private Slack for affiliates. 04 Done-for-you marketing assets. 05 Early access to upcoming courses, launches, and exclusive content. 06 Higher commission tiers available for top affiliates. Potential earnings · Gold tier examples ~$70 per flagship course sale ~$180+ per bundle sale, with higher potential for B2B and team training referrals. One sale is real revenue; many affiliates scale with just a few well-placed links. $18,692.99 in 8 months Shared weekly on YouTube, LinkedIn, newsletters, and learning guides. 66k followers across platforms. “Repeat visibility + multi-channel placement.” $6,730.21 from just 3 posts One technical blog mention and two LinkedIn posts, to 69k followers on a single platform. “Minimal effort, solid return.” 05 Questions", + "heading": "Already have an audience?", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-7fc3ea3f008013d94dc54555", + "text": "Join our referral program and turn your referrals into free courses and 20% revenue share." + }, + { + "span_id": "span-8445d96bf233835e8e8a6101", + "text": "Join the program We respect your privacy." + }, + { + "span_id": "span-a266c2c22fc5c77b9c0aed95", + "text": "Unsubscribe at any time." + }, + { + "span_id": "span-b6efd06d626f7f832356be12", + "text": "03 We do the heavy lifting" + }, + { + "span_id": "span-0e100982717d88f32a1f1ecc", + "text": "You don't need to be a marketing expert to succeed." + }, + { + "span_id": "span-ef08263607f508f075509ab7", + "text": "From banners to social media posts, the creative work is already done for you." + }, + { + "span_id": "span-dc9c40ed4f26b36f57aed580", + "text": "Use pre-written emails to save time and drive clicks." + }, + { + "span_id": "span-e2b0202f1fcbaae6bbc0c538", + "text": "Your first email includes a breakdown of our courses so you can promote what fits your audience best." + }, + { + "span_id": "span-068471f43d50eb768db6f185", + "text": "All assets are tested and designed to help you earn more." + }, + { + "span_id": "span-7514e5bbc50642f7d033d133", + "text": "Get brand assets" + }, + { + "span_id": "span-d1f33880cd3d4ca6f77224f2", + "text": "04 For creators" + }, + { + "span_id": "span-c99a00397869d5c0a0d82899", + "text": "Reply \"Affiliate\" to the welcome email to get enrolled and start earning." + }, + { + "span_id": "span-19e6a2e7df12d208ef07aeb8", + "text": "01 20% commission on every sale." + }, + { + "span_id": "span-9d5187e15562a68cac5dd812", + "text": "Earn up to $180+ per bundle." + }, + { + "span_id": "span-e9db3989bf3df22a602ca72e", + "text": "02 Free access to the courses you promote." + }, + { + "span_id": "span-1727363b9b5311f3340e65da", + "text": "03 Private Slack for affiliates." + }, + { + "span_id": "span-b6589bf8ca5dabd1d0385987", + "text": "04 Done-for-you marketing assets." + }, + { + "span_id": "span-c6156cd4d63c983ea46a8961", + "text": "05 Early access to upcoming courses, launches, and exclusive content." + }, + { + "span_id": "span-027b19245ec60e212274e847", + "text": "06 Higher commission tiers available for top affiliates." + }, + { + "span_id": "span-fc74ab1decd831a387c0febb", + "text": "Potential earnings · Gold tier examples ~$70 per flagship course sale" + }, + { + "span_id": "span-b7a565cd5a748c096586945e", + "text": "~$180+ per bundle sale, with higher potential for B2B and team training referrals." + }, + { + "span_id": "span-435ecea9c472dcd99ea92c39", + "text": "One sale is real revenue; many affiliates scale with just a few well-placed links." + }, + { + "span_id": "span-bd1dbc392af0a0466ec583c4", + "text": "$18,692.99 in 8 months" + }, + { + "span_id": "span-85608764c5e7b7af1b85c07c", + "text": "Shared weekly on YouTube, LinkedIn, newsletters, and learning guides." + }, + { + "span_id": "span-153d0834c0b90b6df4386a7f", + "text": "66k followers across platforms. “Repeat visibility + multi-channel placement.”" + }, + { + "span_id": "span-9d09fb4b59b3f220a71f48ca", + "text": "$6,730.21 from just 3 posts" + }, + { + "span_id": "span-211f9b05b434c45fd156363c", + "text": "One technical blog mention and two LinkedIn posts, to 69k followers on a single platform. “Minimal effort, solid return.”" + }, + { + "span_id": "span-cf19a792157e8613b93021b6", + "text": "05 Questions" + } + ] + }, + { + "chunk_id": "tai-4938511855ab76f05ce7a0d0", + "text": "Frequently asked questions\nHow do I join the affiliate program? + Use the signup form above, and you'll get an email with all the details to get started right away. Do I need an audience to earn? + Not at all. You can start by referring friends casually and earn free rewards. How are referrals tracked? + For affiliates, we use Thinkific's built-in tracking. For casual referrers, we verify manually by email. How do I get paid as an affiliate? + Payouts are made monthly via PayPal. Can I do both? + Yes. Start with referral rewards, and once you hit 10 referrals, we'll invite you to the 20% commission program.\n\nStart earning with the courses you already trust\nStart referring now Get brand assets", + "heading": "Frequently asked questions", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-84fe65512f4d8a4c6507011d", + "text": "Use the signup form above, and you'll get an email with all the details to get started right away." + }, + { + "span_id": "span-dc9471a008e0eb1b3864d8b7", + "text": "Not at all." + }, + { + "span_id": "span-79998d8615d9aea1ea4ac1a7", + "text": "You can start by referring friends casually and earn free rewards." + }, + { + "span_id": "span-51d2ea2e3396e8d9ada6cc6d", + "text": "For affiliates, we use Thinkific's built-in tracking." + }, + { + "span_id": "span-80eb9728e9b4de4684116d9d", + "text": "For casual referrers, we verify manually by email." + }, + { + "span_id": "span-1d5fbdd818ac9f54bb8b928f", + "text": "Payouts are made monthly via PayPal." + }, + { + "span_id": "span-7138d27c80830ca0026a5b3b", + "text": "Start with referral rewards, and once you hit 10 referrals, we'll invite you to the 20% commission program." + }, + { + "span_id": "span-e47fc983ca91e899d4233a0e", + "text": "Start referring now Get brand assets" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "69666855399941a48d9a16d9ce264ddbc934fe5068dc2f3cee44a333752e794b", + "content_hash": "69666855399941a48d9a16d9ce264ddbc934fe5068dc2f3cee44a333752e794b", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/affiliate/" + ], + "evidence_hash": "1794da368d98c9909a4950fa73ad6ef90eac4cc8142b10c74923c0d72166ff8e" + }, + { + "discovered_url": "https://towardsai.com/academy/agent-engineering-free-preview/", + "url": "https://towardsai.com/academy/agent-engineering-free-preview/", + "canonical_url": "https://towardsai.com/academy/agent-engineering-free-preview/", + "host": "towardsai.com", + "path": "/academy/agent-engineering-free-preview", + "kind": "free_resource", + "offer_id": "agent-engineering-free-preview", + "entity_id": "offer:agent-engineering-free-preview", + "title": "Agent Engineering Course Preview · 7 Free Lessons | Towards AI Academy", + "review_title": "Agent Engineering free preview", + "reviewed_summary": "Free preview of Agent Engineering.", + "meta_description": "Preview the Agent Engineering course with 7 full lessons free. Build a research agent and writing workflow connected through MCP, and learn the techniques behind every production agent system.", + "headings": [ + "Agent Engineering Course Preview", + "Inside the Free Preview", + "What's Included", + "7 Full Lessons From the Course", + "5 Code Lessons With Production Patterns", + "2 MCP Agent Prototypes" + ], + "text": "Free preview · 7 full lessons\n\nAgent Engineering Course Preview\nBuild a research agent and writing workflow connected through MCP in 7 free lessons from the course. Build Your First Agent Built for: Software Engineers · Developers · Data Engineers · AI/ML Enthusiasts\n\nInside the Free Preview\nYou'll build on a working prototype of the multi-agent system built across the full course: a research agent and writing workflow connected through MCP. This is a lightweight version designed to show you how the pieces fit together before you build them from scratch. 📩 Want to take this system to production? We'll send you exactly how, watch your inbox and spam.\n\nWhat's Included\n\n7 Full Lessons From the Course\nFull lessons with code you can run: from building your first agent prototypes to the foundational techniques that power every production system.\n\n5 Code Lessons With Production Patterns\nContext engineering, structured outputs, tool calling, chaining, and routing, each built from scratch, which you can reference, reuse, and build on.\n\n2 MCP Agent Prototypes\nA lightweight research agent and writing workflow you can run, inspect, and experiment with. These are starter versions of the production systems you'll build in the full course.", + "links": [ + { + "text": "Build Your First Agent", + "url": "https://academy.towardsai.net/enroll/3406924?et=free_trial" + } + ], + "chunks": [ + { + "chunk_id": "tai-d1a0556060d40457afcc7ffd", + "text": "Free preview · 7 full lessons\n\nAgent Engineering Course Preview\nBuild a research agent and writing workflow connected through MCP in 7 free lessons from the course. Build Your First Agent Built for: Software Engineers · Developers · Data Engineers · AI/ML Enthusiasts\n\nInside the Free Preview\nYou'll build on a working prototype of the multi-agent system built across the full course: a research agent and writing workflow connected through MCP. This is a lightweight version designed to show you how the pieces fit together before you build them from scratch. 📩 Want to take this system to production? We'll send you exactly how, watch your inbox and spam.\n\nWhat's Included\n\n7 Full Lessons From the Course\nFull lessons with code you can run: from building your first agent prototypes to the foundational techniques that power every production system.\n\n5 Code Lessons With Production Patterns\nContext engineering, structured outputs, tool calling, chaining, and routing, each built from scratch, which you can reference, reuse, and build on.\n\n2 MCP Agent Prototypes\nA lightweight research agent and writing workflow you can run, inspect, and experiment with. These are starter versions of the production systems you'll build in the full course.", + "heading": "Inside the Free Preview", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-fcd0edfc2c58b3fed20ac1e3", + "text": "Free preview · 7 full lessons" + }, + { + "span_id": "span-425cf1fbdf4bb03a5d171659", + "text": "Build a research agent and writing workflow connected through MCP in 7 free lessons from the course." + }, + { + "span_id": "span-7371454fcf3fe83c8649fdd8", + "text": "Build Your First Agent" + }, + { + "span_id": "span-23ab989fc58ea774f61fc38a", + "text": "Built for: Software Engineers · Developers · Data Engineers · AI/ML Enthusiasts" + }, + { + "span_id": "span-13c2ffbe83512d68f93da56d", + "text": "You'll build on a working prototype of the multi-agent system built across the full course: a research agent and writing workflow connected through MCP." + }, + { + "span_id": "span-3e754866b401662df5eb2c8b", + "text": "This is a lightweight version designed to show you how the pieces fit together before you build them from scratch." + }, + { + "span_id": "span-84edae386eb50858cbf6482e", + "text": "📩 Want to take this system to production?" + }, + { + "span_id": "span-78c65fd32910347e2fafa99c", + "text": "We'll send you exactly how, watch your inbox and spam." + }, + { + "span_id": "span-9cbec7a3707ecd92cad1b289", + "text": "What's Included" + }, + { + "span_id": "span-9294be259da28b35d230ba8c", + "text": "7 Full Lessons From the Course" + }, + { + "span_id": "span-f3c05c84ed8ca7dac8a5be3a", + "text": "Full lessons with code you can run: from building your first agent prototypes to the foundational techniques that power every production system." + }, + { + "span_id": "span-6cbd052029732390642e56b9", + "text": "5 Code Lessons With Production Patterns" + }, + { + "span_id": "span-a42e81b4baca336786181d30", + "text": "Context engineering, structured outputs, tool calling, chaining, and routing, each built from scratch, which you can reference, reuse, and build on." + }, + { + "span_id": "span-c45754308ac09a619de817ea", + "text": "2 MCP Agent Prototypes" + }, + { + "span_id": "span-bfc1f6146661a1f56f45f122", + "text": "A lightweight research agent and writing workflow you can run, inspect, and experiment with." + }, + { + "span_id": "span-456079288a331afd3cdd0d7d", + "text": "These are starter versions of the production systems you'll build in the full course." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "4c0d86f421fb8697ace1c798d2c8ba7ee63b2cf030bd68918b323df9b239d6fc", + "content_hash": "4c0d86f421fb8697ace1c798d2c8ba7ee63b2cf030bd68918b323df9b239d6fc", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/agent-engineering-free-preview/" + ], + "evidence_hash": "7317197d66c68ce82001d28b86f35229252cd2ca4175603d2b04d770738abacf" + }, + { + "discovered_url": "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/", + "url": "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/", + "canonical_url": "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/", + "host": "towardsai.com", + "path": "/academy/full-stack-ai-engineering-free-preview", + "kind": "free_resource", + "offer_id": "full-stack-ai-engineering-free-preview", + "entity_id": "offer:full-stack-ai-engineering-free-preview", + "title": "Free Preview · Full Stack AI Engineering | Towards AI Academy", + "review_title": "Full Stack AI Engineering free preview", + "reviewed_summary": "Free preview of Full Stack AI Engineering.", + "meta_description": "Try the Full Stack AI Engineering course free and see exactly what you'll learn: RAG, agents, fine-tuning, and deployment, with a certification capstone.", + "headings": [ + "Is this the right upskill for your career?" + ], + "text": "Free preview · Full Stack AI Engineering\n\nIs this the right upskill for your career?\nTry the free preview and see for yourself. You'll see exactly what you'll learn, with insider tips on breaking into AI engineering sent to your inbox. no card required · instant access · watch your inbox (and spam) Free preview lessons · no card required Free preview lessons from the flagship course See exactly what you'll learn inside 92 lessons in the full curriculum Email tips on breaking into AI engineering Instant access, keep the lessons forever Access free lessons Sign up on Towards AI Academy and the lessons open right away", + "links": [ + { + "text": "Access free lessons", + "url": "https://academy.towardsai.net/enroll/2858239?et=free_trial" + } + ], + "chunks": [ + { + "chunk_id": "tai-0374b417d1d45f771275c897", + "text": "Free preview · Full Stack AI Engineering\n\nIs this the right upskill for your career?\nTry the free preview and see for yourself. You'll see exactly what you'll learn, with insider tips on breaking into AI engineering sent to your inbox. no card required · instant access · watch your inbox (and spam) Free preview lessons · no card required Free preview lessons from the flagship course See exactly what you'll learn inside 92 lessons in the full curriculum Email tips on breaking into AI engineering Instant access, keep the lessons forever Access free lessons Sign up on Towards AI Academy and the lessons open right away", + "heading": "Is this the right upskill for your career?", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-ab42103087d74ee41dc9060e", + "text": "Free preview · Full Stack AI Engineering" + }, + { + "span_id": "span-e6f32dbfee4fcaa66e66d54a", + "text": "Try the free preview and see for yourself." + }, + { + "span_id": "span-53da8ea44ccd8e4a60b61051", + "text": "You'll see exactly what you'll learn, with insider tips on breaking into AI engineering sent to your inbox." + }, + { + "span_id": "span-cc59d3cc85c5b410e2b5027a", + "text": "no card required · instant access · watch your inbox (and spam)" + }, + { + "span_id": "span-27ed9bdfc8a0c00a0d6e69b6", + "text": "Free preview lessons · no card required" + }, + { + "span_id": "span-4f7be72215ea9aef38211528", + "text": "Free preview lessons from the flagship course See exactly what you'll learn inside 92 lessons in the full curriculum Email tips on breaking into AI engineering Instant access, keep the lessons forever" + }, + { + "span_id": "span-40022fcb236252c5d625dca0", + "text": "Access free lessons Sign up on Towards AI Academy and the lessons open right away" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "05a98cb6b1d4bde0f22f684ffaf0893dd8100178145cfc60523980512803bc3b", + "content_hash": "05a98cb6b1d4bde0f22f684ffaf0893dd8100178145cfc60523980512803bc3b", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/full-stack-ai-engineering-free-preview/" + ], + "evidence_hash": "025bd31d1695890fcaeb1ffd33e8c2d4cc2d2b445b6598e55df04998c6ff78c4" + }, + { + "discovered_url": "https://towardsai.com/academy/book/", + "url": "https://towardsai.com/academy/book/", + "canonical_url": "https://towardsai.com/academy/book/", + "host": "towardsai.com", + "path": "/academy/book", + "kind": "book", + "offer_id": null, + "entity_id": "page:towardsai.com/academy/book", + "title": "Building AI for Production · The Book: Resources & Links | Towards AI", + "review_title": "The Book", + "reviewed_summary": "Towards AI book page.", + "meta_description": "Every notebook, checkpoint, repository, and reference from Building AI for Production (second edition), chapter by chapter. 10,000+ copies sold.", + "headings": [ + "Building AI for Production", + "Resources & Links", + "Key Updates from Edition 1 to Edition 2", + "A Note on Library and Model Versioning", + "Introduction", + "Book Library Requirements", + "Resources", + "Chapter I: Introduction to LLMs", + "Research Papers", + "Chapter II: LLM Architectures & Landscape", + "Notebook", + "Chapter III: LLM Landscape", + "Chapter IV: Introduction to Prompting", + "Chapter V: Retrieval-Augmented Generation", + "Chapter VI: Introduction to Langchain & LlamaIndex", + "Chapter VII: Prompting with LangChain", + "Chapter VIII: Indexes, Retrievers, and Data Preparation", + "Book File", + "Chapter IX: Advanced RAG", + "Chapter X: Agents", + "Dataset", + "Chapter XI: Fine-Tuning", + "Book Model Checkpoints, Requirements, Datasets, W&B Reports", + "Chapter XII: Deployment", + "Conclusion", + "Further Reading and Courses", + "Previous Courses", + "Free Resources" + ], + "text": "The book · second edition\n\nBuilding AI for Production\nEnhancing LLM abilities and reliability with fine-tuning and RAG. 10,000+ copies sold. This page collects every notebook, checkpoint, repository, and reference from the book, chapter by chapter. Get it in a course bundle Jump to the resources code notebooks · model checkpoints · papers and guides, chapter by chapter\n\nResources & Links\nThis page is a comprehensive compilation of all the links and resources in the book “Building AI for Production: Enhancing LLM Abilities and Reliability with Fine-Tuning and RAG”. Here, you’ll find a collection of code notebooks, checkpoints, GitHub repositories, learning resources, and all other materials shared throughout the book. It is organized chapter-wise and presented in chronological order for easy access. If you see discrepancies between the code in the book and the code in colab, or want to improve the colabs with new updates, please feel free to create a pull request in the GitHub .\n\nKey Updates from Edition 1 to Edition 2\nChapter Update Description Chapter 1 Expanded to include the latest benchmarks, such as LMSYS, and a run-through of recent models like GPT-4 and Gemini 1.5 and technniques like Infinite Attention. Chapter 2 Extended industry applications of LLMs in sectors like media, education, finance, and medicine, with a deeper dive into pecific use cases in each industry, such as autocompletion, code prediction, and debugging in technology and software. Chapter 3 Minor restructuring to improve logical flow and progressive understanding of LLM challenges and solutions. Chapter 4 A new section on prompt injection introduces this emerging security challenge, detailing its types, impact on reliability, with solutions such as guardrails and safeguards to protect LLM integrity. Chapter 5: RAG (Previously, Introduction to LangChain and LlamaIndex) Includes a step-by-step guide to building a basic Retrieval-Augmented Generation (RAG) pipeline from scratch, covering essentials like embeddings, cosine similarity, and vector stores. This foundation equips you to apply modern frameworks like LlamaIndex and LangChain more efficiently or go on your own with custom implementations and prepares you for their evolution better. Chapter 6: Introduction to LangChain & LlamaIndex (Previously Prompting with LangChain) Introduces foundational elements of LangChain as part of a complete system, providing a comprehensive understanding of how each component functions within a broader context. This structured overview acts as a roadmap, enabling a clearer grasp of RAG pipelines in the upcoming chapters. Chapter 7: Prompting with LangChain (Previously RAG) Includes LangChain Chains, previously part of the RAG chapter, for clarity. Chapter 8: Indexes, Retrievers, and Data preparation (New Chapter) Indexes, Retrievers, and Data Preparation are essential components of a RAG pipeline. While these concepts were introduced in the first edition, this updated edition includes a dedicated chapter that focuses on their foundational principles. This approach ensures that readers can effectively scale LLM applications, optimize performance, and enhance response quality. Additionally, by emphasizing the fundamentals, this edition allows readers to understand and implement RAG concepts independently, without relying exclusively on frameworks like LangChain. Chapter 9: Advanced RAG Only structural updates Chapter 10: Agents Only structural updates Chapter 11: Fine-tuning Only structural updates Chapter 12: Deployment and Optimization The updated version takes a deeper dive into essential techniques for LLM deployment and optimization, making it more practical and relevant for current AI development needs. For example, the book explores model distillation , a powerful technique to reduce inference costs and improve latency, with a detailed case study on Google’s Gemma 2 , demonstrating its real-world impact. With open-source LLMs growing in popularity, this edition also covers the deployment of LLMs on various cloud platforms, including Together AI, Groq, Fireworks AI, and Replicate. This broader approach helps readers find cost-effective and scalable solutions for real-world applications.\n\nA Note on Library and Model Versioning\nLLMs are advancing rapidly, but the core skills and tools covered in this book—like fine-tuning, prompt engineering, and retrieval-augmented generation— will remain essential for adapting next-generation models to specific data, workflows, and industries. These principles will stay relevant across models, even as some specific libraries evolve. For seamless code execution, we’ve included a requirements file for library versions. If you’re running notebooks on Google Colab, be aware that libraries like “pytorch” and “Transformers” are pre-installed. Should compatibility issues arise, try uninstalling these libraries in Colab and reinstalling the specified versions from the requirements file. Switching to newer LLMs is straightforward. For instance, with OpenAI models, you can update the model simply by changing its name in the code. We recommend using GPT-4o-mini over “GPT-3.5 Turbo” in the book examples. Regularly checking documentation for Langchain, LlamaIndex, and OpenAI is also encouraged to stay aligned with updates and best practices. This approach ensures your skills remain applicable in the dynamic LLM field.\n\nA Note on Library and Model Versioning\n\nIntroduction\nNo Notebooks.\n\nBook Library Requirements\nRequirements\n\nResources\nPython & Other Technical Notes: Our guide to starting in AI (Python, Math, and more resources. All free) Towards AI Open Source AI chatbot: AI Tutor Discord Community: Learn AI Together Coding Environment and Packages: Visual Studio Code\n\nChapter I: Introduction to LLMs\nNo Notebooks.\n\nResearch Papers\nAttention Is All You Need (Section: Transformers) Training Compute-Optimal Large Language Models . (Section: Scaling Laws) Emergent Abilities of Large Language Models (Section: What are Emergent Abilities) Evaluation Benchmarks for Emergent Abilities : Massive Multi-task Language Understanding MMLU | Word in Context Optimization Techniques to Expand the Context Window: ALiBi Positional Encoding | Sparse Attention | FlashAttention | Multi-Query Attention (MQA) FlashAttention-2 LONGNET: Scaling Transformers to 1,000,000,000 Tokens . A Survey of Large Language Models (Section: A Timeline of the Most Popular LLMs) Evaluation Benchmarks for Emergent Abilities GitHub Repo : BIG-Bench suite | TruthfulQA |\n\nChapter II: LLM Architectures & Landscape\n\nNotebook\nUnderstanding Transformer (Section: The Architecture in Action) Transformer Architecture (Section: Transformer Model’s Design Choices) Resources & Additional Links Tutorial: Let’s build GPT: from scratch, in code, spelled out Demo Environment: IDEFICS Playground GitHub Repo: minGPT – A PyTorch re-implementation of GPT, both training and inference Open AI Blog Post: InstructGPT LLM Leaderboard: Chatbot Arena LLaVA – An Instruction-tuned LMM: Vicuna Open Flamingo Beyond Vision and Language (Models): PandaGPT | ImageBind | SpeechGPT | NExT-GPT Proprietary and Open-Source LLMs: Cohere LLMs | Open AI GPT 3.5 | Anthropic’s Claude Models | Google Deepmind’s Gemini | Meta’s Llama Models | Falcon | Dolly | Open Assistant | Mistral LLMs Research Paper: “Vision Transformers: An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale” Research Paper: “Multimodal Foundation Models: From Specialists to General-Purpose Assistants” Research Paper: Flamingo: a Visual Language Model for Few-Shot Learning\n\nChapter III: LLM Landscape\nNo Notebooks. Research Papers: Evaluating LLM Performance (Benchmarks) The GLUE Benchmark The SuperGLUE Benchmark The BIG-Bench Benchmark The HELM Benchmark The FLASK Benchmark\n\nChapter IV: Introduction to Prompting\n\nNotebook\nNo Notebooks. Resources Prompting A Hands-on Guide to Prompt Engineering with ChatGPT and GPT-3 Prompt Engineering Tips and Tricks with GPT-3 Prompt Engineering LLMs with LangChain and W&B\n\nChapter V: Retrieval-Augmented Generation\n\nNotebook\nBuilding a Basic RAG Pipeline from Scratch (Section: Building a Basic RAG Pipeline from Scratch) Resources Research Paper: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Blog: Retrieval Augmented Generation (RAG) Deep Lake Vector Store\n\nChapter VI: Introduction to Langchain & LlamaIndex\n\nNotebook\nBuilding Application Powered by LLMs with LangChain (Section: Tutorial 1: Building LLM-Powered Applications with LangChain) Build a News Articles Summarizer (Section: Tutorial 2: Building a News Articles Summarizer) LlamaIndex Introduction (Section: LlamaIndex Introduction) Resources LangChain Documentation Useful LangChain Components: | Prompts | Output Parsers | Retrievers | Document Loaders | Text Splitters | Indexes | Embeddings models | Vector Stores | Agents | Chain | Tool | Memory | Callbacks Useful LangChain Agents: Zero-shot ReAct | Structured Input ReAct | OpenAI Functions Agent | Self-Ask with Search Agent | ReAct Document Store Agent | Plan-and-execute agents The output for the Building LLM-Powered Applications with LangChain section is based on The One Page Linux Manual: A summary of useful Linux commands Building applications with LLMs through composability A Complete Guide to LangChain: Building Powerful Applications with LLMs LlamaIndex Index Guide LlamaIndex: How to use Index correctly Defining a Custom Query Engine Working Example of Implementing Routers LlamaIndex documentation Financial Document Analysis with LlamaIndex LlamaIndex: Adding Personal Data to LLMs Llamahub\n\nChapter VII: Prompting with LangChain\n\nNotebook\nUsing Prompt Templates (Section: What are LangChain Prompt Templates) Getting the Best of Few Shot Prompts & Example Selectors (Section: Few Shot Prompts and Example Selectors) Chains and Why They Are Used Notebook (Section: What are LangChain Chains) Managing Outputs with Output Parsers (Section: Tutorial 1: Managing Outputs with Output Parsers) Improving Our News Articles Summarizer (Section: Tutorial 2: Improving Our News Articles Summarizer) Creating Knowledge Graphs from Textual Data Unveiling Hidden Connections (Section: Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections) Resources Building a Knowledge Base from Texts: a Full Practical Example GitHub: langchain Knowledge Graph Visualization: NetworkX library Knowledge Graph Visualization: Pyvis library\n\nChapter VIII: Indexes, Retrievers, and Data Preparation\n\nNotebook\nWhat are Text Splitters and Why They are Useful (Section: Text Splitters) Create a YouTube Video Summarizer Using Whisper and LangChain (Section: Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain) Guarding Against Undesirable Outputs With the Self-Critique Chain (Section: Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain) Guarding Against Undesirable Outputs With the Self-Critique Chain Example (Section: Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot)\n\nBook File\nBook File Requirements Sample PDF Used for Customizing Text Splitters Example: The One Page Linux Manual Tokens and APIs & Packages Ffmpeg Open AI API Keys Eleven API Keys Activeloop Token Resources Tutorial: How to install ffmpeg Documentation: Text wrapping and filling Code: idontcalculate/langchain GitHub Repo: JarvisBase LangChain Docs: Split by character | Split code | Recursively split by character | Summarization Open AI Whisper Acitveloop Docs: Deep Lake Vector Store in LangChain Documentation: Self-critique chain with constitutional AI\n\nChapter IX: Advanced RAG\n\nNotebook\nMasterting Advanced RAG (Section: Using a Query Engine to Answer Queries) RAG Metrics & Evaluations (Section: RAG Metrics & Evaluation) LangSmith Introduction (Section: LangChain LangSmith and LangChain Hub) Resources Source Document for the Query Engine Example: Text Data for the example (Section: Query Engine Example) Tutorial: Building an Advanced Fusion Retriever from Scratch LangChain Docs: Query construction Cohere Reranking Tutorial: Hands-on Tutorial for Implementing Small-to-Big Retrieval Colab Notebook: Cohere Rerank Endpoint Blog: Complex Query Resolution through LlamaIndex Utilizing Recursive Retrieval, Document Agents, and Sub Question Query Decomposition Blog: Improving Retrieval Performance by Fine-tuning Cohere Reranker with LlamaIndex LlamaIndex Notebook LllamaIndex Docs (Section: The Role of the Retrieval Step): Retriever Query Engine with Custom Retrievers – Simple Hybrid Search | Metadata Filtering | Cohere Rerank | Document Summary Index LllamaIndex Docs (Section: RAG Metrics): Correctness | Faithfulness | Context Relevancy | Guideline Adherence | Embedding Semantic Similari | Finetuning Embeddings | Evaluating – LlamaIndex | Retrieval Evaluation | Golden Dataset | Response Evaluation | Recursive Retriever + Query Engine Demo | Query Engine | Chat Engine Creating the Dataset openai-cookbook RAGAS GitHub repository RagEvaluatorPack Downloading a LlamaDataset from LlamaHub LangSmith Hub-examples: LangSmith cookbook The Art of LangSmith LangServe Github Repository\n\nChapter X: Agents\n\nNotebook\nUsing AutoGPT with LangChain (Section: Using AutoGPT with LangChain) Using AutoGPT with LangChain Output (Section: Using AutoGPT with LangChain) Building Autonomous Agents to Create Analysis Reports (Section: Tutorial 1: Building Agents for Analysis Report Creation) Query and Summarize a DB with LlamaIndex (Section: Tutorial 2: Query and Summarize a DB with LlamaIndex) Building Agents with OpenAI Assistants (Section: Tutorial 3: Building Agents with OpenAI Assistants) Multimodal Finance + Deep Memory (Section: Tutorial 5: Multimodal Financial Document Analysis from PDFs)\n\nDataset\nDataset for the Multimodal Financial Document Analysis Example: Tesla Q3 Financial Report Preprocessed Text/Label for the Multimodal Financial Document Analysis Preprocessed Graphs for the Multimodal Financial Document Analysis Resources GitHub: Babyagi Inspired Projects GitHub: Agent Simulations GitHub: CAMEL Role-Playing Autonomous Cooperative Agents GitHub; LlamaHub LlamaIndex Docs: Data Agents | OpenAI Agent with Query Engine Tools | Multi-Document Agents Blog: OpenAI Assistants API: Walk-through and Coding a Research Assistant GitHub: HuggingFace Inference Community Colab Notebook: Assistants API OpenAI Docs: OpenAI Knowledge Retrieval Blog: Function Calling OpenAI GitHub: LangChain OpenGPTs Blog: Maximizing LangChain Efficiency: Agents and ReAct Method Review LangChain Docs: Defining Custom Tools Tutorial: Installing Poppler on Windows Tutorial: Installing Tesseract on Windows Website: AutoGPT Website: BabyAGI Research: On AutoGPT – LessWrong Website: CAMEL Research Paper: The CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society paper Research Paper: Generative Agents: Interactive Simulacra of Human Behavior Website: OpenGPTs\n\nChapter XI: Fine-Tuning\n\nNotebook\nFineTuning a LLM Lima CPU (Section: Tutorial 1: SFT with LoRA) FineTuning a LLM Financial Sentiment CPU (Section: Tutorial 2: Using SFT and LoRA for Financial Sentiment) Create a Dataset For Cohere Fine-Tuning (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data) Fine-Tuning Using Cohere for Medical Data (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data) Finetuning a LLM QloRA (Section: Tutorial 4/Supervised Fine-Tuning Notebook) Finetuning a Reward Model (Section: Tutorial 4/Training a Reward Model Notebook) Finetune RLHF (Section: Tutorial 4/RLHF)\n\nBook Model Checkpoints, Requirements, Datasets, W&B Reports\nOPT fine-tuned LIMA checkpoint on CPU (Section: Practical Example: SFT with LoRA) OPT Fine-tuned finGPT with CPU (Section: Using SFT for Financial Sentiment) The Merged Model Checkpoint (2GB) (Section: Supervised Fine-Tuning Notebook) Requirements (Section: Supervised Fine-Tuning Notebook) The Reward Model Checkpoint (Step 1000 – 2GB) (Section: Training a Reward Model Notebook) Requirements (Section: Training a Reward Model Notebook) The Merged RL Model Checkpoint (2GB) (Section: RLHF) Requirements (Section: RLHF) BC5CDR Dataset in JSON format (Section: Fine-Tuning a Cohere LLM with Medical Data) Preprocessed Dataset (Section: Fine-Tuning a Cohere LLM with Medical Data) Complete Dataset (Section: Supervised Fine-Tuning Notebook) OpenOrca Dataset (Section: Supervised Fine-Tuning Notebook & Section: RLHF) “helpfulness/harmless”: (hh) by Anthropic (Section: Training a Reward Model Notebook) OPT Fine-tuned LIMA CPU (Section: Practical Example: SFT with LoRA) Weights & Bias Report (Section: Supervised Fine-Tuning Notebook) Weights & Bias Report (Section: Training a Reward Model Notebook) Weights and Biases report (Section: RLHF) Resources Research Paper: Low-Rank Adaptation (LoRA) Research Paper: QLoRA: An Efficient Variant of LoRA Open-source Resources for LoRA: PEFT Library | Lit-GPT Cohere Docs: Fine-tuning an Embedding Model for Classification Research Paper: Reinforcement Learning from Human Feedback Research Paper: LIMA: Less Is More for Alignment Research Paper: Direct Preference Optimization (DPO) Research Paper: Google DeepMind’s Reinforced Self-Training (ReST) Research Paper: Reinforcement Learning from AI Feedback (RLAIF)\n\nChapter XII: Deployment\n\nNotebook\nBenchmark Inference (Section: Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP)) Resources Research Paper: Model Compression Research Paper: Distilling the Knowledge in a Neural Network Research Paper: A Survey of Quantization Methods for Efficient Neural Network Inference Research Paper: Sparsity in Deep Learning GitHub: Hugging Face Optimum GitHub: Intel Neural Compressor Research Paper: LLM.int8(): 8bit Matrix Multiplication for Transformers at Scale Research Paper: GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers Research Paper: AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration Research Paper: Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures Reasearch Paper: A Simple and Effective Pruning Approach for Large Language Models Reasearch Paper: Structured Pruning of Deep Convolutional Neural Networks Research Paper: The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks A complete list of tasks supported with Simple Quantization (Using CLI) The Docker Image under Latitude: Llama 2 API Inference\n\nConclusion\nNo Notebooks.\n\nFurther Reading and Courses\nTowards AI Open Source AI-Tutor HuggingFace: AI Tutor Chatbot\n\nPrevious Courses\nTraining & Fine-Tuning LLMs for Production LangChain & Vector Databases in Production\n\nFree Resources\nTowards AI Newsletter What’s AI YouTube Channel Towards AI Academy Note: This webpage has been updated to follow the order and structure of the second edition. All resources from the first edition are still available but may have been rearranged to match the new sequence. Additionally, you’ll find new links and resources exclusive to the second edition—an added benefit for readers with the first edition.", + "links": [ + { + "text": "Get it in a course bundle", + "url": "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/" + }, + { + "text": "the GitHub", + "url": "https://github.com/towardsai/ragbook-notebooks" + }, + { + "text": "Google’s Gemma 2", + "url": "https://arxiv.org/pdf/2408.00118" + }, + { + "text": "GPT-4o-mini", + "url": "https://platform.openai.com/docs/models/gpt-4o-mini" + }, + { + "text": "Requirements", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/book_requirements.txt" + }, + { + "text": "Our guide to starting in AI (Python, Math, and more resources. All free)", + "url": "https://github.com/louisfb01/start-machine-learning" + }, + { + "text": "AI Tutor", + "url": "https://aitutor.towardsai.com/" + }, + { + "text": "Learn AI Together", + "url": "https://discord.gg/learnaitogether?ref=louisbouchard.ai" + }, + { + "text": "Visual Studio Code", + "url": "https://code.visualstudio.com/" + }, + { + "text": "Attention Is All You Need", + "url": "https://arxiv.org/pdf/1706.03762.pdf" + }, + { + "text": "Training Compute-Optimal Large Language Models", + "url": "https://arxiv.org/abs/2203.15556" + }, + { + "text": "Emergent Abilities of Large Language Models", + "url": "https://arxiv.org/pdf/2206.07682.pdf" + }, + { + "text": "Massive Multi-task Language Understanding", + "url": "https://arxiv.org/abs/2009.03300" + }, + { + "text": "MMLU", + "url": "https://arxiv.org/abs/2009.03300" + }, + { + "text": "Word in Context", + "url": "https://arxiv.org/abs/1808.09121v3" + }, + { + "text": "ALiBi Positional Encoding", + "url": "https://arxiv.org/abs/2108.12409" + }, + { + "text": "Sparse Attention", + "url": "https://ai.googleblog.com/2021/03/constructing-transformers-for-longer.html" + }, + { + "text": "FlashAttention", + "url": "https://arxiv.org/abs/2205.14135" + }, + { + "text": "Multi-Query Attention (MQA)", + "url": "https://arxiv.org/pdf/1911.02150.pdf" + }, + { + "text": "FlashAttention-2", + "url": "https://crfm.stanford.edu/2023/07/17/flash2.html?utm_content=bufferca8a7&utm_medium=social&utm_source=linkedin.com&utm_campaign=buffer" + }, + { + "text": "LONGNET: Scaling Transformers to 1,000,000,000 Tokens", + "url": "https://arxiv.org/pdf/2307.02486.pdf" + }, + { + "text": "A Survey of Large Language Models", + "url": "https://arxiv.org/pdf/2303.18223.pdf" + }, + { + "text": "BIG-Bench suite", + "url": "https://github.com/google/BIG-bench/blob/main/bigbench/benchmark_tasks/README.md" + }, + { + "text": "TruthfulQA", + "url": "https://github.com/sylinrl/TruthfulQA" + }, + { + "text": "Understanding Transformer", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2002%20-%20Understanding_Transformer.ipynb" + }, + { + "text": "Transformer Architecture", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2002%20-%20Transformers_Architectures.ipynb" + }, + { + "text": "Let’s build GPT: from scratch, in code, spelled out", + "url": "https://www.youtube.com/watch?v=kCc8FmEb1nY" + }, + { + "text": "IDEFICS Playground", + "url": "https://huggingface.co/spaces/HuggingFaceM4/idefics_playground" + }, + { + "text": "minGPT – A PyTorch re-implementation of GPT, both training and inference", + "url": "https://github.com/karpathy/minGPT" + }, + { + "text": "InstructGPT", + "url": "https://openai.com/research/instruction-following" + }, + { + "text": "Chatbot Arena", + "url": "https://lmsys.org/blog/2023-05-03-arena/" + }, + { + "text": "Vicuna", + "url": "https://lmsys.org/blog/2023-03-30-vicuna/" + }, + { + "text": "Open Flamingo", + "url": "https://github.com/mlfoundations/open_flamingo" + }, + { + "text": "PandaGPT", + "url": "https://github.com/yxuansu/PandaGPT" + }, + { + "text": "ImageBind", + "url": "https://imagebind.metademolab.com/" + }, + { + "text": "SpeechGPT", + "url": "https://github.com/0nutation/SpeechGPT" + }, + { + "text": "NExT-GPT", + "url": "https://arxiv.org/pdf/2309.05519.pdf" + }, + { + "text": "Cohere LLMs", + "url": "https://dashboard.cohere.com/welcome/register?ref=txt.cohere.com" + }, + { + "text": "Open AI GPT 3.5", + "url": "https://platform.openai.com/docs/models/overview" + }, + { + "text": "Anthropic’s Claude Models", + "url": "https://docs.anthropic.com/claude/docs/models-overview" + }, + { + "text": "Google Deepmind’s Gemini", + "url": "https://deepmind.google/technologies/gemini/" + }, + { + "text": "Meta’s Llama Models", + "url": "https://llama.meta.com/llama2/" + }, + { + "text": "Falcon", + "url": "https://falconllm.tii.ae/falcon-models.html" + }, + { + "text": "Dolly", + "url": "https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm" + }, + { + "text": "Open Assistant", + "url": "https://open-assistant.io/bye" + }, + { + "text": "Mistral LLMs", + "url": "https://docs.mistral.ai/" + }, + { + "text": "“Vision Transformers: An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale”", + "url": "https://arxiv.org/abs/2010.11929" + }, + { + "text": "“Multimodal Foundation Models: From Specialists to General-Purpose Assistants”", + "url": "https://arxiv.org/pdf/2309.10020.pdf" + }, + { + "text": "Flamingo: a Visual Language Model for Few-Shot Learning", + "url": "https://arxiv.org/abs/2204.14198" + }, + { + "text": "The GLUE Benchmark", + "url": "https://gluebenchmark.com/" + }, + { + "text": "The SuperGLUE Benchmark", + "url": "https://super.gluebenchmark.com/" + }, + { + "text": "The BIG-Bench Benchmark", + "url": "https://github.com/google/BIG-bench" + }, + { + "text": "The HELM Benchmark", + "url": "https://crfm.stanford.edu/2022/11/17/helm.html" + }, + { + "text": "The FLASK Benchmark", + "url": "https://arxiv.org/abs/2307.10928" + }, + { + "text": "Prompting", + "url": "https://www.louisbouchard.ai/prompting-explained/" + }, + { + "text": "A Hands-on Guide to Prompt Engineering with ChatGPT and GPT-3", + "url": "https://dev.to/mmz001/a-hands-on-guide-to-prompt-engineering-with-chatgpt-and-gpt-3-4127" + }, + { + "text": "Prompt Engineering Tips and Tricks with GPT-3", + "url": "https://blog.andrewcantino.com/blog/2021/04/21/prompt-engineering-tips-and-tricks/" + }, + { + "text": "Prompt Engineering LLMs with LangChain and W&B", + "url": "https://wandb.ai/a-sh0ts/langchain_callback_demo/reports/Prompt-Engineering-LLMs-with-LangChain-and-W-B--VmlldzozNjk1NTUw" + }, + { + "text": "Building a Basic RAG Pipeline from Scratch", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2005%20-%20Building%20a%20basic%20RAG%20pipeline%20%20from%20scratch.ipynb" + }, + { + "text": "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks", + "url": "https://arxiv.org/abs/2005.11401v4" + }, + { + "text": "Retrieval Augmented Generation (RAG)", + "url": "https://www.louisbouchard.ai/rag-explained/" + }, + { + "text": "Deep Lake Vector Store", + "url": "https://docs.activeloop.ai/" + }, + { + "text": "Building Application Powered by LLMs with LangChain", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2005%20-%20Building_Applications_Powered_by_LLMs_with_LangChain.ipynb" + }, + { + "text": "Build a News Articles Summarizer", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2005%20-%20Build_a_News_Articles_Summarizer.ipynb" + }, + { + "text": "LlamaIndex Introduction", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2005%20-%20LlamaIndex_Introduction.ipynb" + }, + { + "text": "LangChain Documentation", + "url": "https://python.langchain.com/docs/get_started/introduction/" + }, + { + "text": "Prompts", + "url": "https://python.langchain.com/docs/modules/model_io/prompts/" + }, + { + "text": "Output Parsers", + "url": "https://python.langchain.com/docs/modules/model_io/output_parsers/" + }, + { + "text": "Retrievers", + "url": "https://python.langchain.com/docs/modules/data_connection/retrievers/" + }, + { + "text": "Document Loaders", + "url": "https://python.langchain.com/docs/modules/data_connection/document_loaders/" + }, + { + "text": "Text Splitters", + "url": "https://python.langchain.com/docs/modules/data_connection/document_transformers/" + }, + { + "text": "Indexes", + "url": "https://python.langchain.com/docs/modules/data_connection/indexing/" + }, + { + "text": "Embeddings models", + "url": "https://python.langchain.com/docs/modules/data_connection/text_embedding/" + }, + { + "text": "Vector Stores", + "url": "https://python.langchain.com/docs/modules/data_connection/vectorstores/" + }, + { + "text": "Agents", + "url": "https://python.langchain.com/docs/modules/agents/" + }, + { + "text": "Chain", + "url": "https://python.langchain.com/docs/modules/chains/" + }, + { + "text": "Tool", + "url": "https://python.langchain.com/docs/modules/agents/tools/" + }, + { + "text": "Memory", + "url": "https://python.langchain.com/docs/modules/memory/" + }, + { + "text": "Callbacks", + "url": "https://python.langchain.com/docs/modules/callbacks/" + }, + { + "text": "Zero-shot ReAct", + "url": "https://python.langchain.com/docs/modules/agents/agent_types/react" + }, + { + "text": "Structured Input ReAct", + "url": "https://python.langchain.com/docs/modules/agents/agent_types/structured_chat" + }, + { + "text": "OpenAI Functions Agent", + "url": "https://python.langchain.com/docs/modules/agents/agent_types/openai_functions_agent" + }, + { + "text": "Self-Ask with Search Agent", + "url": "https://python.langchain.com/docs/modules/agents/agent_types/self_ask_with_search" + }, + { + "text": "ReAct Document Store Agent", + "url": "https://python.langchain.com/docs/modules/agents/agent_types/react" + }, + { + "text": "Plan-and-execute agents", + "url": "https://blog.langchain.dev/plan-and-execute-agents/" + }, + { + "text": "The One Page Linux Manual: A summary of useful Linux commands", + "url": "https://www.cheat-sheets.org/saved-copy/The%20One%20Page%20Linux%20Manual.pdf" + }, + { + "text": "Building applications with LLMs through composability", + "url": "https://pypi.org/project/langchain/" + }, + { + "text": "A Complete Guide to LangChain: Building Powerful Applications with LLMs", + "url": "https://notes.aimodels.fyi/a-complete-guide-to-langchain-building-powerful-applications-with-large-language-models/" + }, + { + "text": "LlamaIndex Index Guide", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/indexing/index_guide/" + }, + { + "text": "LlamaIndex: How to use Index correctly", + "url": "https://howaibuildthis.substack.com/p/llamaindex-how-to-use-index-correctly" + }, + { + "text": "Defining a Custom Query Engine", + "url": "https://docs.llamaindex.ai/en/stable/examples/query_engine/custom_query_engine/" + }, + { + "text": "Working Example of Implementing Routers", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/querying/router/" + }, + { + "text": "LlamaIndex documentation", + "url": "https://gpt-index.readthedocs.io/en/stable/" + }, + { + "text": "Financial Document Analysis with LlamaIndex", + "url": "https://cookbook.openai.com/examples/third_party/financial_document_analysis_with_llamaindex" + }, + { + "text": "LlamaIndex: Adding Personal Data to LLMs", + "url": "https://www.datacamp.com/tutorial/llama-index-adding-personal-data-to-llms" + }, + { + "text": "Llamahub", + "url": "https://llamahub.ai/" + }, + { + "text": "Using Prompt Templates", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2006%20-%20Using_Prompt_Templates.ipynb" + }, + { + "text": "Getting the Best of Few Shot Prompts & Example Selectors", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2006%20-%20Getting_the_Best_of_Few_Shot_Prompts_and_Example_Selectors.ipynb" + }, + { + "text": "Chains and Why They Are Used Notebook", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2007%20-%20Chains_and_Why_They_Are_Used.ipynb" + }, + { + "text": "Managing Outputs with Output Parsers", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2006%20-%20Managing_Outputs_with_Output_Parsers.ipynb" + }, + { + "text": "Improving Our News Articles Summarizer", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2006%20-%20Improving_Our_News_Articles_Summarizer.ipynb" + }, + { + "text": "Creating Knowledge Graphs from Textual Data Unveiling Hidden Connections", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2006%20-%20Creating_Knowledge_Graphs_from_Textual_Data_Unveiling_Hidden_Connections.ipynb" + }, + { + "text": "Building a Knowledge Base from Texts: a Full Practical Example", + "url": "https://medium.com/nlplanet/building-a-knowledge-base-from-texts-a-full-practical-example-8dbbffb912fa" + }, + { + "text": "GitHub: langchain", + "url": "https://github.com/langchain-ai/langchain/blob/master/libs/langchain/langchain/indexes/prompts/knowledge_triplet_extraction.py" + }, + { + "text": "NetworkX library", + "url": "https://networkx.org/" + }, + { + "text": "Pyvis library", + "url": "https://pyvis.readthedocs.io/en/latest/" + }, + { + "text": "What are Text Splitters and Why They are Useful", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2007%20-%20What_are_Text_Splitters_and_Why_They_are_Useful_.ipynb" + }, + { + "text": "Create a YouTube Video Summarizer Using Whisper and LangChain", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2007%20-%20Create_a_YouTube_Video_Summarizer_Using_Whisper_and_LangChain_.ipynb" + }, + { + "text": "Guarding Against Undesirable Outputs With the Self-Critique Chain", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2007%20-%20Guarding_Against_Undesirable_Outputs_with_the_Self_Critique_Chain.ipynb" + }, + { + "text": "Guarding Against Undesirable Outputs With the Self-Critique Chain Example", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2007%20-%20Guarding_Against_Undesirable_Outputs_with_the_Self_Critique_Chain_Example.ipynb" + }, + { + "text": "Book File Requirements", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/va_requirements.txt" + }, + { + "text": "The One Page Linux Manual", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/The%20One%20Page%20Linux%20Manual.pdf" + }, + { + "text": "Ffmpeg", + "url": "https://ffmpeg.org/" + }, + { + "text": "Open AI API Keys", + "url": "https://platform.openai.com/account/api-keys" + }, + { + "text": "Eleven API Keys", + "url": "https://elevenlabs.io/" + }, + { + "text": "Activeloop Token", + "url": "https://www.activeloop.ai/" + }, + { + "text": "How to install ffmpeg", + "url": "https://www.hostinger.com/tutorials/how-to-install-ffmpeg" + }, + { + "text": "Text wrapping and filling", + "url": "https://docs.python.org/3/library/textwrap.html" + }, + { + "text": "idontcalculate/langchain", + "url": "https://github.com/idontcalculate/langchain/blob/main/yt-whisper-sum.ipynb" + }, + { + "text": "JarvisBase", + "url": "https://github.com/peterw/JarvisBase" + }, + { + "text": "Split by character", + "url": "https://python.langchain.com/docs/modules/data_connection/document_transformers/character_text_splitter" + }, + { + "text": "Split code", + "url": "https://python.langchain.com/docs/modules/data_connection/document_transformers/code_splitter" + }, + { + "text": "Recursively split by character", + "url": "https://python.langchain.com/docs/modules/data_connection/document_transformers/recursive_text_splitter" + }, + { + "text": "Summarization", + "url": "https://python.langchain.com/docs/use_cases/summarization" + }, + { + "text": "Open AI Whisper", + "url": "https://openai.com/research/whisper" + }, + { + "text": "Deep Lake Vector Store in LangChain", + "url": "https://docs.activeloop.ai/example-code/tutorials/vector-store/deep-lake-vector-store-in-langchain" + }, + { + "text": "Self-critique chain with constitutional AI", + "url": "https://python.langchain.com/docs/modules/chains" + }, + { + "text": "Masterting Advanced RAG", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2008%20-%20Mastering_Advanced_RAG.ipynb" + }, + { + "text": "RAG Metrics & Evaluations", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2008%20-%20RAG_Metrics%26Evaluation.ipynb" + }, + { + "text": "LangSmith Introduction", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2008%20-%20LangSmith_Introduction.ipynb" + }, + { + "text": "Text Data for the example", + "url": "https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt" + }, + { + "text": "Building an Advanced Fusion Retriever from Scratch", + "url": "https://docs.llamaindex.ai/en/stable/examples/low_level/fusion_retriever.html" + }, + { + "text": "Query construction", + "url": "https://blog.langchain.dev/query-construction/" + }, + { + "text": "Reranking", + "url": "https://cohere.com/rerank?ref=txt.cohere.com" + }, + { + "text": "Hands-on Tutorial for Implementing Small-to-Big Retrieval", + "url": "https://docs.llamaindex.ai/en/stable/examples/node_postprocessor/MetadataReplacementDemo/" + }, + { + "text": "Cohere Rerank Endpoint", + "url": "https://colab.research.google.com/drive/1PW3_MLYBnalsomyy2Ia-KvEq7Fayp0W6" + }, + { + "text": "Complex Query Resolution through LlamaIndex Utilizing Recursive Retrieval, Document Agents, and Sub Question Query Decomposition", + "url": "https://medium.com/@sauravjoshi23/complex-query-resolution-through-llamaindex-utilizing-recursive-retrieval-document-agents-and-sub-d4861ecd54e6" + }, + { + "text": "Improving Retrieval Performance by Fine-tuning Cohere Reranker with LlamaIndex", + "url": "https://blog.llamaindex.ai/improving-retrieval-performance-by-fine-tuning-cohere-reranker-with-llamaindex-16c0c1f9b33b" + }, + { + "text": "LlamaIndex Notebook", + "url": "https://colab.research.google.com/github/jerryjliu/llama_index/blob/main/docs/examples/customization/streaming/SimpleIndexDemo-streaming.ipynb" + }, + { + "text": "Retriever Query Engine with Custom Retrievers – Simple Hybrid Search", + "url": "https://docs.llamaindex.ai/en/stable/examples/query_engine/CustomRetrievers.html" + }, + { + "text": "Metadata Filtering", + "url": "https://docs.llamaindex.ai/en/stable/examples/vector_stores/WeaviateIndexDemo/" + }, + { + "text": "Cohere Rerank", + "url": "https://docs.llamaindex.ai/en/stable/examples/node_postprocessor/CohereRerank/" + }, + { + "text": "Document Summary Index", + "url": "https://docs.llamaindex.ai/en/stable/examples/index_structs/doc_summary/DocSummary.html" + }, + { + "text": "Correctness", + "url": "https://docs.llamaindex.ai/en/latest/examples/evaluation/correctness_eval.html" + }, + { + "text": "Faithfulness", + "url": "https://docs.llamaindex.ai/en/latest/examples/evaluation/faithfulness_eval.html" + }, + { + "text": "Context Relevancy", + "url": "https://docs.llamaindex.ai/en/latest/examples/evaluation/relevancy_eval.html" + }, + { + "text": "Guideline Adherence", + "url": "https://docs.llamaindex.ai/en/latest/examples/evaluation/guideline_eval.html" + }, + { + "text": "Embedding Semantic Similari", + "url": "https://docs.llamaindex.ai/en/latest/examples/evaluation/semantic_similarity_eval.html" + }, + { + "text": "Finetuning Embeddings", + "url": "https://docs.llamaindex.ai/en/stable/optimizing/fine-tuning/fine-tuning/" + }, + { + "text": "Evaluating – LlamaIndex", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/evaluating/root.html" + }, + { + "text": "Retrieval Evaluation", + "url": "https://docs.llamaindex.ai/en/stable/examples/evaluation/retrieval/retriever_eval/" + }, + { + "text": "Golden Dataset", + "url": "https://github.com/microsoft/promptflow-resource-hub/blob/main/sample_gallery/golden_dataset/copilot-golden-dataset-creation-guidance.md" + }, + { + "text": "Response Evaluation", + "url": "https://docs.llamaindex.ai/en/latest/module_guides/evaluating/usage_pattern.html" + }, + { + "text": "Recursive Retriever + Query Engine Demo", + "url": "https://docs.llamaindex.ai/en/stable/examples/query_engine/pdf_tables/recursive_retriever/" + }, + { + "text": "Query Engine", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/deploying/query_engine/root.html" + }, + { + "text": "Chat Engine", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/deploying/chat_engines/root.html" + }, + { + "text": "Creating the Dataset", + "url": "https://github.com/microsoft/promptflow-resource-hub/blob/main/sample_gallery/golden_dataset/copilot-golden-dataset-creation-guidance.md" + }, + { + "text": "openai-cookbook", + "url": "https://github.com/openai/openai-cookbook/blob/main/examples/evaluation/Evaluate_RAG_with_LlamaIndex.ipynb" + }, + { + "text": "RAGAS", + "url": "https://github.com/explodinggradients/ragas/blob/main/docs/howtos/integrations/llamaindex.ipynb" + }, + { + "text": "Downloading a LlamaDataset from LlamaHub", + "url": "https://docs.llamaindex.ai/en/stable/examples/llama_dataset/downloading_llama_datasets.html" + }, + { + "text": "LangSmith", + "url": "https://python.langchain.com/docs/langsmith" + }, + { + "text": "Hub-examples: LangSmith cookbook", + "url": "https://github.com/langchain-ai/langsmith-cookbook/blob/main/hub-examples/retrieval-qa-chain/retrieval-qa.ipynb" + }, + { + "text": "The Art of LangSmith", + "url": "https://betterprogramming.pub/the-art-of-langsmithing-42dcd191a220" + }, + { + "text": "LangServe Github Repository", + "url": "https://betterprogramming.pub/the-art-of-langsmithing-42dcd191a220" + }, + { + "text": "Using AutoGPT with LangChain", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20Using_AutoGPT_with_LangChain.ipynb" + }, + { + "text": "Using AutoGPT with LangChain Output", + "url": "https://github.com/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20Using_AutoGPT_with_LangChain.ipynb" + }, + { + "text": "Building Autonomous Agents to Create Analysis Reports", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20Building%20Agents%20for%20Analysis%20Report%20Creation.ipynb" + }, + { + "text": "Query and Summarize a DB with LlamaIndex", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20Query%20and%20Zummarize%20a%20DB%20with%20LlamaIndex.ipynb" + }, + { + "text": "Building Agents with OpenAI Assistants", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20Building%20Agents%20with%20OpenAI%20Assistants.ipynb" + }, + { + "text": "Multimodal Finance + Deep Memory", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2009%20-%20MultiModal_Fincance_%2B_DeepMemory.ipynb" + }, + { + "text": "Tesla Q3 Financial Report", + "url": "https://digitalassets.tesla.com/tesla-contents/image/upload/IR/TSLA-Q3-2023-Update-3.pdf" + }, + { + "text": "Preprocessed Text/Label", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/categorized_elements.pkl" + }, + { + "text": "Preprocessed Graphs", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/graphs_description.pkl" + }, + { + "text": "Babyagi Inspired Projects", + "url": "https://github.com/yoheinakajima/babyagi/blob/main/docs/inspired-projects.md" + }, + { + "text": "Agent Simulations", + "url": "https://github.com/langchain-ai/langchain/blob/master/cookbook/gymnasium_agent_simulation.ipynb" + }, + { + "text": "CAMEL Role-Playing Autonomous Cooperative Agents", + "url": "https://github.com/langchain-ai/langchain/blob/master/cookbook/camel_role_playing.ipynb" + }, + { + "text": "LlamaHub", + "url": "https://github.com/run-llama/llama-hub/tree/main/llama_hub/tools/notebooks" + }, + { + "text": "Data Agents", + "url": "https://docs.llamaindex.ai/en/stable/module_guides/deploying/agents/root.html" + }, + { + "text": "OpenAI Agent with Query Engine Tools", + "url": "https://docs.llamaindex.ai/en/latest/examples/agent/openai_agent_with_query_engine.html" + }, + { + "text": "Multi-Document Agents", + "url": "https://docs.llamaindex.ai/en/latest/examples/agent/multi_document_agents.html" + }, + { + "text": "OpenAI Assistants API: Walk-through and Coding a Research Assistant", + "url": "https://medium.com/@datadrifters/openai-assistants-api-walk-through-and-coding-a-research-assistant-3ddb3d767e99" + }, + { + "text": "HuggingFace Inference Community", + "url": "https://github.com/huggingface/api-inference-community/" + }, + { + "text": "Assistants API", + "url": "https://colab.research.google.com/drive/1GkjoTZUd2rpwIvxmCZreKo5ZUZN6i7vJ" + }, + { + "text": "OpenAI Knowledge Retrieval", + "url": "https://platform.openai.com/docs/assistants/tools/knowledge-retrieval" + }, + { + "text": "Function Calling OpenAI", + "url": "https://tmmtt.medium.com/functions-with-openai-assistant-api-47e1481a0b42" + }, + { + "text": "LangChain OpenGPTs", + "url": "https://github.com/langchain-ai/opengpts" + }, + { + "text": "Maximizing LangChain Efficiency: Agents and ReAct Method Review", + "url": "https://pub.towardsai.com/maximizing-langchain-efficiency-agents-and-react-method-review-3abcf8a5c874" + }, + { + "text": "Defining Custom Tools", + "url": "https://python.langchain.com/docs/modules/agents/tools/custom_tools" + }, + { + "text": "Installing Poppler on Windows", + "url": "https://towardsdatascience.com/poppler-on-windows-179af0e50150" + }, + { + "text": "Installing Tesseract on Windows", + "url": "https://indiantechwarrior.com/how-to-install-tesseract-on-windows/" + }, + { + "text": "AutoGPT", + "url": "https://autogpt.net/" + }, + { + "text": "BabyAGI", + "url": "https://babyagi.org/" + }, + { + "text": "On AutoGPT – LessWrong", + "url": "https://www.lesswrong.com/posts/566kBoPi76t8KAkoD/on-autogpt" + }, + { + "text": "CAMEL", + "url": "https://www.camel-ai.org/" + }, + { + "text": "The CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society paper", + "url": "https://ghli.org/camel.pdf" + }, + { + "text": "Generative Agents: Interactive Simulacra of Human Behavior", + "url": "https://arxiv.org/pdf/2304.03442.pdf" + }, + { + "text": "OpenGPTs", + "url": "https://opengpts-example-vz4y4ooboq-uc.a.run.app/" + }, + { + "text": "FineTuning a LLM Lima CPU", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20FineTuning_a_LLM_LIMA_CPU.ipynb" + }, + { + "text": "FineTuning a LLM Financial Sentiment CPU", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20FineTuning_a_LLM_Financial_Sentiment_CPU.ipynb" + }, + { + "text": "Create a Dataset For Cohere Fine-Tuning", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20Create_Dataset_For_Cohere_Fine_Tuning.ipynb" + }, + { + "text": "Fine-Tuning Using Cohere for Medical Data", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20Fine_Tuning_using_Cohere_for_Medical_Data.ipynb" + }, + { + "text": "Finetuning a LLM QloRA", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20FineTuning_a_LLM_QLoRA.ipynb" + }, + { + "text": "Finetuning a Reward Model", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20FineTuning_Reward_Model.ipynb" + }, + { + "text": "Finetune RLHF", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2010%20-%20FineTune_RLHF.ipynb" + }, + { + "text": "OPT fine-tuned LIMA checkpoint on CPU", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/OPT-fine_tuned-LIMA-CPU.zip" + }, + { + "text": "OPT Fine-tuned finGPT with CPU", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/OPT-fine_tuned-FinGPT-CPU.zip" + }, + { + "text": "The Merged Model Checkpoint (2GB)", + "url": "https://drive.google.com/file/d/1D9rH2kLiBgRR31xvelOcW09uHzrO5Fbv/view?usp=drive_link" + }, + { + "text": "Requirements", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/requirements-fine-tune.txt" + }, + { + "text": "The Reward Model Checkpoint (Step 1000 – 2GB)", + "url": "https://drive.google.com/file/d/1GWL-ayeeXDCMuqYjvDzgPJiRFmBY6X6Z/view?usp=drive_link" + }, + { + "text": "Requirements", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/requirements-reward.txt" + }, + { + "text": "The Merged RL Model Checkpoint (2GB)", + "url": "https://drive.google.com/file/d/12ekdlETljGrZm_SP50us3XRXIKEMn0C-/view?usp=drive_link" + }, + { + "text": "Requirements", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/requirements-rl.txt" + }, + { + "text": "BC5CDR", + "url": "https://paperswithcode.com/dataset/bc5cdr" + }, + { + "text": "JSON format", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/bc5cdr.json" + }, + { + "text": "Preprocessed Dataset", + "url": "https://github.com/towardsai/rag-ebook-files/blob/main/both_rel_instruct_all.jsonl" + }, + { + "text": "Complete Dataset", + "url": "https://app.activeloop.ai/genai360/OpenOrca-4M/" + }, + { + "text": "OpenOrca Dataset", + "url": "https://huggingface.co/datasets/Open-Orca/OpenOrca" + }, + { + "text": "“helpfulness/harmless”: (hh) by Anthropic", + "url": "https://github.com/anthropics/hh-rlhf" + }, + { + "text": "OPT Fine-tuned LIMA CPU", + "url": "https://wandb.ai/ala_/GenAI360/runs/uhe0kbku?workspace=user-ala_" + }, + { + "text": "Weights & Bias Report", + "url": "https://wandb.ai/ala_/GenAI360/runs/n6czwaqq?workspace=user-ala_" + }, + { + "text": "Weights and Biases report", + "url": "https://wandb.ai/ala_/GenAI360/runs/e9y58bdi?workspace=user-ala_" + }, + { + "text": "Low-Rank Adaptation (LoRA)", + "url": "https://arxiv.org/abs/2106.09685" + }, + { + "text": "QLoRA: An Efficient Variant of LoRA", + "url": "https://arxiv.org/abs/2305.14314" + }, + { + "text": "PEFT Library", + "url": "https://github.com/huggingface/peft" + }, + { + "text": "Lit-GPT", + "url": "https://github.com/Lightning-AI/lit-gpt" + }, + { + "text": "Fine-tuning an Embedding Model for Classification", + "url": "https://docs.cohere.com/docs/finetuning" + }, + { + "text": "Reinforcement Learning from Human Feedback", + "url": "https://arxiv.org/abs/2305.18438" + }, + { + "text": "LIMA: Less Is More for Alignment", + "url": "https://arxiv.org/abs/2305.11206" + }, + { + "text": "Direct Preference Optimization (DPO)", + "url": "https://arxiv.org/pdf/2305.18290.pdf" + }, + { + "text": "Google DeepMind’s Reinforced Self-Training (ReST)", + "url": "https://arxiv.org/pdf/2308.08998.pdf" + }, + { + "text": "Reinforcement Learning from AI Feedback (RLAIF)", + "url": "https://arxiv.org/abs/2212.08073" + }, + { + "text": "Benchmark Inference", + "url": "https://colab.research.google.com/github/towardsai/ragbook-notebooks/blob/main/notebooks/Chapter%2011%20-%20Benchmark_Inference.ipynb" + }, + { + "text": "Model Compression", + "url": "https://www.cs.cornell.edu/~caruana/compression.kdd06.pdf" + }, + { + "text": "Distilling the Knowledge in a Neural Network", + "url": "https://arxiv.org/abs/1503.02531" + }, + { + "text": "A Survey of Quantization Methods for Efficient Neural Network Inference", + "url": "https://arxiv.org/pdf/2103.13630" + }, + { + "text": "Sparsity in Deep Learning", + "url": "https://arxiv.org/abs/2102.00554" + }, + { + "text": "Hugging Face Optimum", + "url": "https://github.com/huggingface/optimum" + }, + { + "text": "Intel Neural Compressor", + "url": "https://github.com/intel/neural-compressor/tree/master" + }, + { + "text": "LLM.int8(): 8bit Matrix Multiplication for Transformers at Scale", + "url": "https://arxiv.org/abs/2208.07339" + }, + { + "text": "GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers", + "url": "https://arxiv.org/abs/2210.17323" + }, + { + "text": "AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration", + "url": "https://arxiv.org/abs/2306.00978" + }, + { + "text": "Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures", + "url": "https://arxiv.org/abs/1607.03250" + }, + { + "text": "A Simple and Effective Pruning Approach for Large Language Models", + "url": "https://arxiv.org/abs/2306.11695" + }, + { + "text": "Structured Pruning of Deep Convolutional Neural Networks", + "url": "https://arxiv.org/abs/1512.08571" + }, + { + "text": "The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks", + "url": "https://arxiv.org/abs/1803.03635" + }, + { + "text": "list of tasks supported", + "url": "https://github.com/huggingface/optimum/blob/3ffebf994fbd579ab9acb589edc401fa66413928/optimum/exporters/tasks.py" + }, + { + "text": "Llama 2 API Inference", + "url": "https://hub.docker.com/r/latitudesh/llama2-api-inference" + }, + { + "text": "Towards AI Open Source AI-Tutor", + "url": "https://aitutor.towardsai.com/" + }, + { + "text": "HuggingFace: AI Tutor Chatbot", + "url": "https://huggingface.co/spaces/towardsai-buster/ai-tutor-chatbot" + }, + { + "text": "Training & Fine-Tuning LLMs for Production", + "url": "https://learn.activeloop.ai/courses/llms" + }, + { + "text": "LangChain & Vector Databases in Production", + "url": "https://learn.activeloop.ai/courses/langchain" + }, + { + "text": "Towards AI Newsletter", + "url": "https://newsletter.towardsai.com/" + }, + { + "text": "What’s AI YouTube Channel", + "url": "https://www.youtube.com/channel/UCUzGQrN-lyyc0BWTYoJM_Sg" + }, + { + "text": "Towards AI Academy", + "url": "https://towardsai.com/towards-ai-academy" + } + ], + "chunks": [ + { + "chunk_id": "tai-54d84cfcf96edc6052cc2322", + "text": "The book · second edition\n\nBuilding AI for Production\nEnhancing LLM abilities and reliability with fine-tuning and RAG. 10,000+ copies sold. This page collects every notebook, checkpoint, repository, and reference from the book, chapter by chapter. Get it in a course bundle Jump to the resources code notebooks · model checkpoints · papers and guides, chapter by chapter\n\nResources & Links\nThis page is a comprehensive compilation of all the links and resources in the book “Building AI for Production: Enhancing LLM Abilities and Reliability with Fine-Tuning and RAG”. Here, you’ll find a collection of code notebooks, checkpoints, GitHub repositories, learning resources, and all other materials shared throughout the book. It is organized chapter-wise and presented in chronological order for easy access. If you see discrepancies between the code in the book and the code in colab, or want to improve the colabs with new updates, please feel free to create a pull request in the GitHub .", + "heading": "Resources & Links", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-7dc71c0ea2eb5191ff2836dc", + "text": "The book · second edition" + }, + { + "span_id": "span-0bda9e88280b760c3134b0cd", + "text": "Enhancing LLM abilities and reliability with fine-tuning and RAG." + }, + { + "span_id": "span-886d215573b97374c8d3f129", + "text": "10,000+ copies sold." + }, + { + "span_id": "span-677be483386c0905c7128b4a", + "text": "This page collects every notebook, checkpoint, repository, and reference from the book, chapter by chapter." + }, + { + "span_id": "span-c79fc759caed67954d9fa52b", + "text": "Get it in a course bundle Jump to the resources" + }, + { + "span_id": "span-fce06d4a591d3c54deafdbb8", + "text": "code notebooks · model checkpoints · papers and guides, chapter by chapter" + }, + { + "span_id": "span-65efcafbc41655122d5f4227", + "text": "This page is a comprehensive compilation of all the links and resources in the book “Building AI for Production: Enhancing LLM Abilities and Reliability with Fine-Tuning and RAG”." + }, + { + "span_id": "span-e11805e679a9a11d1aaaef01", + "text": "Here, you’ll find a collection of code notebooks, checkpoints, GitHub repositories, learning resources, and all other materials shared throughout the book." + }, + { + "span_id": "span-56030070fbd7e75fcbc70b78", + "text": "It is organized chapter-wise and presented in chronological order for easy access." + }, + { + "span_id": "span-a336daa41db13e299d67ab24", + "text": "If you see discrepancies between the code in the book and the code in colab, or want to improve the colabs with new updates, please feel free to create a pull request in the GitHub ." + } + ] + }, + { + "chunk_id": "tai-afe18e81c5bc73adbd8cb1ba", + "text": "Key Updates from Edition 1 to Edition 2\nChapter Update Description Chapter 1 Expanded to include the latest benchmarks, such as LMSYS, and a run-through of recent models like GPT-4 and Gemini 1.5 and technniques like Infinite Attention. Chapter 2 Extended industry applications of LLMs in sectors like media, education, finance, and medicine, with a deeper dive into pecific use cases in each industry, such as autocompletion, code prediction, and debugging in technology and software. Chapter 3 Minor restructuring to improve logical flow and progressive understanding of LLM challenges and solutions. Chapter 4 A new section on prompt injection introduces this emerging security challenge, detailing its types, impact on reliability, with solutions such as guardrails and safeguards to protect LLM integrity. Chapter 5: RAG (Previously, Introduction to LangChain and LlamaIndex) Includes a step-by-step guide to building a basic Retrieval-Augmented Generation (RAG) pipeline from scratch, covering essentials like embeddings, cosine similarity, and vector stores. This foundation equips you to apply modern frameworks like LlamaIndex and LangChain more efficiently or go on your own with custom implementations and prepares you for their evolution better. Chapter 6: Introduction to LangChain & LlamaIndex (Previously Prompting with LangChain) Introduces foundational elements of LangChain as part of a complete system, providing a comprehensive understanding of how each component functions within a broader context. This structured overview acts as a roadmap, enabling a clearer grasp of RAG pipelines in the upcoming chapters. Chapter 7: Prompting with LangChain (Previously RAG) Includes LangChain Chains, previously part of the RAG chapter, for clarity. Chapter 8: Indexes, Retrievers, and", + "heading": "Key Updates from Edition 1 to Edition 2", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-b61f32418984099c613a08e4", + "text": "Key Updates from Edition 1 to Edition 2" + }, + { + "span_id": "span-f73ac095a1ce9abac67b1125", + "text": "Chapter Update Description" + }, + { + "span_id": "span-9928bc884d33b1e0e7d1effb", + "text": "Chapter 1 Expanded to include the latest benchmarks, such as LMSYS, and a run-through of recent models like GPT-4 and Gemini 1.5 and technniques like Infinite Attention." + }, + { + "span_id": "span-395e497fdb44ddc8b686a476", + "text": "Chapter 2 Extended industry applications of LLMs in sectors like media, education, finance, and medicine, with a deeper dive into pecific use cases in each industry, such as autocompletion, code prediction, and debugging in technology and software." + }, + { + "span_id": "span-eff33a0b5860a1b8f3be4fcd", + "text": "Chapter 3 Minor restructuring to improve logical flow and progressive understanding of LLM challenges and solutions." + }, + { + "span_id": "span-7d91ee99af84aa393b4a5b80", + "text": "Chapter 4 A new section on prompt injection introduces this emerging security challenge, detailing its types, impact on reliability, with solutions such as guardrails and safeguards to protect LLM integrity." + }, + { + "span_id": "span-785b0551a4b781e6979622c1", + "text": "Chapter 5: RAG (Previously, Introduction to LangChain and LlamaIndex) Includes a step-by-step guide to building a basic Retrieval-Augmented Generation (RAG) pipeline from scratch, covering essentials like embeddings, cosine similarity, and vector stores." + }, + { + "span_id": "span-66a26a954cfbc7f8b0dd6c66", + "text": "This foundation equips you to apply modern frameworks like LlamaIndex and LangChain more efficiently or go on your own with custom implementations and prepares you for their evolution better." + }, + { + "span_id": "span-6e3190e3fed8c2da764145d5", + "text": "Chapter 6: Introduction to LangChain & LlamaIndex (Previously Prompting with LangChain) Introduces foundational elements of LangChain as part of a complete system, providing a comprehensive understanding of how each component functions within a broader context." + }, + { + "span_id": "span-55019cd846caf0eff6b4dcd5", + "text": "This structured overview acts as a roadmap, enabling a clearer grasp of RAG pipelines in the upcoming chapters." + }, + { + "span_id": "span-7aeb0d991bd101337cdb6e4d", + "text": "Chapter 7: Prompting with LangChain (Previously RAG) Includes LangChain Chains, previously part of the RAG chapter, for clarity." + } + ] + }, + { + "chunk_id": "tai-e3f6190db902d6711b8566a7", + "text": "Key Updates from Edition 1 to Edition 2\nData preparation (New Chapter) Indexes, Retrievers, and Data Preparation are essential components of a RAG pipeline. While these concepts were introduced in the first edition, this updated edition includes a dedicated chapter that focuses on their foundational principles. This approach ensures that readers can effectively scale LLM applications, optimize performance, and enhance response quality. Additionally, by emphasizing the fundamentals, this edition allows readers to understand and implement RAG concepts independently, without relying exclusively on frameworks like LangChain. Chapter 9: Advanced RAG Only structural updates Chapter 10: Agents Only structural updates Chapter 11: Fine-tuning Only structural updates Chapter 12: Deployment and Optimization The updated version takes a deeper dive into essential techniques for LLM deployment and optimization, making it more practical and relevant for current AI development needs. For example, the book explores model distillation , a powerful technique to reduce inference costs and improve latency, with a detailed case study on Google’s Gemma 2 , demonstrating its real-world impact. With open-source LLMs growing in popularity, this edition also covers the deployment of LLMs on various cloud platforms, including Together AI, Groq, Fireworks AI, and Replicate. This broader approach helps readers find cost-effective and scalable solutions for real-world applications.", + "heading": "Key Updates from Edition 1 to Edition 2", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-2c8e65d643ef63eac5a3eb23", + "text": "While these concepts were introduced in the first edition, this updated edition includes a dedicated chapter that focuses on their foundational principles." + }, + { + "span_id": "span-9f6b4adb1ba56ec32aea8a25", + "text": "This approach ensures that readers can effectively scale LLM applications, optimize performance, and enhance response quality." + }, + { + "span_id": "span-7c093edc87a42eaf83cb22be", + "text": "Additionally, by emphasizing the fundamentals, this edition allows readers to understand and implement RAG concepts independently, without relying exclusively on frameworks like LangChain." + }, + { + "span_id": "span-8bf1e098c93e60803f9ce8f0", + "text": "Chapter 9: Advanced RAG Only structural updates" + }, + { + "span_id": "span-bd268e24cceb6eaa743a19ad", + "text": "Chapter 10: Agents Only structural updates" + }, + { + "span_id": "span-5ab551f6855c617bdb876a29", + "text": "Chapter 11: Fine-tuning Only structural updates" + }, + { + "span_id": "span-a9aac75c04f7ef34fb21e08c", + "text": "Chapter 12: Deployment and Optimization The updated version takes a deeper dive into essential techniques for LLM deployment and optimization, making it more practical and relevant for current AI development needs." + }, + { + "span_id": "span-05a37bacac778f3617b4f586", + "text": "For example, the book explores model distillation , a powerful technique to reduce inference costs and improve latency, with a detailed case study on Google’s Gemma 2 , demonstrating its real-world impact." + }, + { + "span_id": "span-e4a335ec7cd6611fa5f6caa9", + "text": "With open-source LLMs growing in popularity, this edition also covers the deployment of LLMs on various cloud platforms, including Together AI, Groq, Fireworks AI, and Replicate." + }, + { + "span_id": "span-f0b6ed45d8c297d305503e30", + "text": "This broader approach helps readers find cost-effective and scalable solutions for real-world applications." + } + ] + }, + { + "chunk_id": "tai-c4df0bd99d66601f8f473d6d", + "text": "A Note on Library and Model Versioning\nLLMs are advancing rapidly, but the core skills and tools covered in this book—like fine-tuning, prompt engineering, and retrieval-augmented generation— will remain essential for adapting next-generation models to specific data, workflows, and industries. These principles will stay relevant across models, even as some specific libraries evolve. For seamless code execution, we’ve included a requirements file for library versions. If you’re running notebooks on Google Colab, be aware that libraries like “pytorch” and “Transformers” are pre-installed. Should compatibility issues arise, try uninstalling these libraries in Colab and reinstalling the specified versions from the requirements file. Switching to newer LLMs is straightforward. For instance, with OpenAI models, you can update the model simply by changing its name in the code. We recommend using GPT-4o-mini over “GPT-3.5 Turbo” in the book examples. Regularly checking documentation for Langchain, LlamaIndex, and OpenAI is also encouraged to stay aligned with updates and best practices. This approach ensures your skills remain applicable in the dynamic LLM field.\n\nA Note on Library and Model Versioning\n\nIntroduction\nNo Notebooks.\n\nBook Library Requirements\nRequirements\n\nResources\nPython & Other Technical Notes: Our guide to starting in AI (Python, Math, and more resources. All free) Towards AI Open Source AI chatbot: AI Tutor Discord Community: Learn AI Together Coding Environment and Packages: Visual Studio Code\n\nChapter I: Introduction to LLMs\nNo Notebooks.", + "heading": "A Note on Library and Model Versioning", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-750f5738c5e7ca24f469967f", + "text": "LLMs are advancing rapidly, but the core skills and tools covered in this book—like fine-tuning, prompt engineering, and retrieval-augmented generation— will remain essential for adapting next-generation models to specific data, workflows, and industries." + }, + { + "span_id": "span-e0d56c74ae29776dbdc4dd65", + "text": "These principles will stay relevant across models, even as some specific libraries evolve." + }, + { + "span_id": "span-59dd4e1407805dfb639a3a98", + "text": "For seamless code execution, we’ve included a requirements file for library versions." + }, + { + "span_id": "span-65cf56e2f19a17d324ce5b76", + "text": "If you’re running notebooks on Google Colab, be aware that libraries like “pytorch” and “Transformers” are pre-installed." + }, + { + "span_id": "span-a2b9d1b1eaff694bc7ddde1d", + "text": "Should compatibility issues arise, try uninstalling these libraries in Colab and reinstalling the specified versions from the requirements file." + }, + { + "span_id": "span-cd69056adf36bb0ec9a6cc80", + "text": "Switching to newer LLMs is straightforward." + }, + { + "span_id": "span-75398738668e9169c4f19f42", + "text": "For instance, with OpenAI models, you can update the model simply by changing its name in the code." + }, + { + "span_id": "span-6c313685423818bba7750c33", + "text": "We recommend using GPT-4o-mini over “GPT-3.5 Turbo” in the book examples." + }, + { + "span_id": "span-ecd802b71878032bb77d6d3b", + "text": "Regularly checking documentation for Langchain, LlamaIndex, and OpenAI is also encouraged to stay aligned with updates and best practices." + }, + { + "span_id": "span-df0c33c1eeb32030f9f548c6", + "text": "This approach ensures your skills remain applicable in the dynamic LLM field." + }, + { + "span_id": "span-ae216197a3804c0d3c9ac46f", + "text": "A Note on Library and Model Versioning" + }, + { + "span_id": "span-a0bb3bbae3d562883f3867a2", + "text": "No Notebooks." + }, + { + "span_id": "span-0bd99266665e6b7ea887f0a9", + "text": "Book Library Requirements" + }, + { + "span_id": "span-aaedff35ab769c7992fa842e", + "text": "Python & Other Technical Notes: Our guide to starting in AI (Python, Math, and more resources." + }, + { + "span_id": "span-0fc58c8a683d717b8582ab1d", + "text": "All free)" + }, + { + "span_id": "span-6273fc73d33c1035aceaa2d8", + "text": "Towards AI Open Source AI chatbot: AI Tutor" + }, + { + "span_id": "span-527b52a4b05ca07445d1cca8", + "text": "Discord Community: Learn AI Together" + }, + { + "span_id": "span-3cf4e1e4047a0dd90c42bd8e", + "text": "Coding Environment and Packages: Visual Studio Code" + } + ] + }, + { + "chunk_id": "tai-b996e5555079b92013a24c3e", + "text": "Research Papers\nAttention Is All You Need (Section: Transformers) Training Compute-Optimal Large Language Models . (Section: Scaling Laws) Emergent Abilities of Large Language Models (Section: What are Emergent Abilities) Evaluation Benchmarks for Emergent Abilities : Massive Multi-task Language Understanding MMLU | Word in Context Optimization Techniques to Expand the Context Window: ALiBi Positional Encoding | Sparse Attention | FlashAttention | Multi-Query Attention (MQA) FlashAttention-2 LONGNET: Scaling Transformers to 1,000,000,000 Tokens . A Survey of Large Language Models (Section: A Timeline of the Most Popular LLMs) Evaluation Benchmarks for Emergent Abilities GitHub Repo : BIG-Bench suite | TruthfulQA |\n\nChapter II: LLM Architectures & Landscape\n\nNotebook\nUnderstanding Transformer (Section: The Architecture in Action) Transformer Architecture (Section: Transformer Model’s Design Choices) Resources & Additional Links Tutorial: Let’s build GPT: from scratch, in code, spelled out Demo Environment: IDEFICS Playground GitHub Repo: minGPT – A PyTorch re-implementation of GPT, both training and inference Open AI Blog Post: InstructGPT LLM Leaderboard: Chatbot Arena LLaVA – An Instruction-tuned LMM: Vicuna Open Flamingo Beyond Vision and Language (Models): PandaGPT | ImageBind | SpeechGPT | NExT-GPT Proprietary and Open-Source LLMs: Cohere LLMs | Open AI GPT 3.5 | Anthropic’s Claude Models | Google Deepmind’s Gemini | Meta’s Llama Models | Falcon | Dolly | Open Assistant | Mistral LLMs Research Paper: “Vision Transformers: An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale” Research Paper: “Multimodal Foundation Models: From Specialists to General-Purpose Assistants” Research Paper: Flamingo: a Visual Language Model for Few-Shot Learning", + "heading": "Notebook", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-9afc9fa6316a02efd81c6d98", + "text": "Attention Is All You Need (Section: Transformers)" + }, + { + "span_id": "span-1ed44270138bb1f74f2b90c3", + "text": "Training Compute-Optimal Large Language Models ." + }, + { + "span_id": "span-73fb41707fbdffe81d544539", + "text": "(Section: Scaling Laws)" + }, + { + "span_id": "span-f1660faf64afb38e67cb9ee9", + "text": "Emergent Abilities of Large Language Models (Section: What are Emergent Abilities)" + }, + { + "span_id": "span-1f9d090ad2ade82b4424efd6", + "text": "Evaluation Benchmarks for Emergent Abilities : Massive Multi-task Language Understanding MMLU | Word in Context" + }, + { + "span_id": "span-73f9252f860c3bb1ff01b0a8", + "text": "Optimization Techniques to Expand the Context Window: ALiBi Positional Encoding | Sparse Attention | FlashAttention | Multi-Query Attention (MQA)" + }, + { + "span_id": "span-828695f98b8b38dcc7e9a30c", + "text": "FlashAttention-2" + }, + { + "span_id": "span-b48c050338164b8856530dec", + "text": "LONGNET: Scaling Transformers to 1,000,000,000 Tokens ." + }, + { + "span_id": "span-f02a668605a78ce507d74961", + "text": "A Survey of Large Language Models (Section: A Timeline of the Most Popular LLMs)" + }, + { + "span_id": "span-1b1703ca1bd009c159384f5f", + "text": "Evaluation Benchmarks for Emergent Abilities GitHub Repo : BIG-Bench suite | TruthfulQA |" + }, + { + "span_id": "span-9eb12630dec83fca7f1404a0", + "text": "Chapter II: LLM Architectures & Landscape" + }, + { + "span_id": "span-128550490df7743f070068e2", + "text": "Understanding Transformer (Section: The Architecture in Action)" + }, + { + "span_id": "span-a178dc0dd3d7c3517b6428eb", + "text": "Transformer Architecture (Section: Transformer Model’s Design Choices)" + }, + { + "span_id": "span-3aa5fed4533315bbfc0a205d", + "text": "Resources & Additional Links" + }, + { + "span_id": "span-35866fce9b40543bf1fd6bd2", + "text": "Tutorial: Let’s build GPT: from scratch, in code, spelled out" + }, + { + "span_id": "span-1e5c7844f86ee8e0ee4b2bd1", + "text": "Demo Environment: IDEFICS Playground" + }, + { + "span_id": "span-69d18a0213e885fc6b5be6ad", + "text": "GitHub Repo: minGPT – A PyTorch re-implementation of GPT, both training and inference" + }, + { + "span_id": "span-1b854888ec2fa33ecf956273", + "text": "Open AI Blog Post: InstructGPT" + }, + { + "span_id": "span-227b24c37891df5565fb8083", + "text": "LLM Leaderboard: Chatbot Arena" + }, + { + "span_id": "span-eea00e53cc51d915dbab0046", + "text": "LLaVA – An Instruction-tuned LMM: Vicuna" + }, + { + "span_id": "span-8d25fe431692789c1e497739", + "text": "Open Flamingo" + }, + { + "span_id": "span-bb90a91b91effcd324ef3e74", + "text": "Beyond Vision and Language (Models): PandaGPT | ImageBind | SpeechGPT | NExT-GPT" + }, + { + "span_id": "span-d5715774ef721d4696959a30", + "text": "Proprietary and Open-Source LLMs: Cohere LLMs | Open AI GPT 3.5 | Anthropic’s Claude Models | Google Deepmind’s Gemini | Meta’s Llama Models | Falcon | Dolly | Open Assistant | Mistral LLMs" + }, + { + "span_id": "span-85e05673dbffbc464f38beb1", + "text": "Research Paper: “Vision Transformers: An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale”" + }, + { + "span_id": "span-71bbbfef4e87cb5a286d958b", + "text": "Research Paper: “Multimodal Foundation Models: From Specialists to General-Purpose Assistants”" + }, + { + "span_id": "span-a10858cfb04b1e1512433939", + "text": "Research Paper: Flamingo: a Visual Language Model for Few-Shot Learning" + } + ] + }, + { + "chunk_id": "tai-50e5936f98043dd1c36674bd", + "text": "Chapter III: LLM Landscape\nNo Notebooks. Research Papers: Evaluating LLM Performance (Benchmarks) The GLUE Benchmark The SuperGLUE Benchmark The BIG-Bench Benchmark The HELM Benchmark The FLASK Benchmark\n\nChapter IV: Introduction to Prompting\n\nNotebook\nNo Notebooks. Resources Prompting A Hands-on Guide to Prompt Engineering with ChatGPT and GPT-3 Prompt Engineering Tips and Tricks with GPT-3 Prompt Engineering LLMs with LangChain and W&B\n\nChapter V: Retrieval-Augmented Generation\n\nNotebook\nBuilding a Basic RAG Pipeline from Scratch (Section: Building a Basic RAG Pipeline from Scratch) Resources Research Paper: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks Blog: Retrieval Augmented Generation (RAG) Deep Lake Vector Store\n\nChapter VI: Introduction to Langchain & LlamaIndex", + "heading": "Notebook", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-4348c1c641f009b1d7cdf9ba", + "text": "No Notebooks." + }, + { + "span_id": "span-576e0623ec21a66754028ef9", + "text": "Research Papers: Evaluating LLM Performance (Benchmarks)" + }, + { + "span_id": "span-07ad8a8fe00e67921695fcd9", + "text": "The GLUE Benchmark" + }, + { + "span_id": "span-5713289ceb0c5ecd8d7ad9d5", + "text": "The SuperGLUE Benchmark" + }, + { + "span_id": "span-79c7f32bd9c81777e865e592", + "text": "The BIG-Bench Benchmark" + }, + { + "span_id": "span-cd57622ae9591e1986282e83", + "text": "The HELM Benchmark" + }, + { + "span_id": "span-d1d554fb40511bd72165bc35", + "text": "The FLASK Benchmark" + }, + { + "span_id": "span-d4c93335ec2e5abfe5ebfc38", + "text": "Chapter IV: Introduction to Prompting" + }, + { + "span_id": "span-32275f39aed37b77b1fb6b45", + "text": "A Hands-on Guide to Prompt Engineering with ChatGPT and GPT-3" + }, + { + "span_id": "span-c922e59ded56448f3f2842bc", + "text": "Prompt Engineering Tips and Tricks with GPT-3" + }, + { + "span_id": "span-233189e86c035dc382367ceb", + "text": "Prompt Engineering LLMs with LangChain and W&B" + }, + { + "span_id": "span-4f8234093be4f5da104e1d94", + "text": "Chapter V: Retrieval-Augmented Generation" + }, + { + "span_id": "span-8cc43b6e2084c06192dc1039", + "text": "Building a Basic RAG Pipeline from Scratch (Section: Building a Basic RAG Pipeline from Scratch)" + }, + { + "span_id": "span-f0627c07f7b4caa5bc7db6ca", + "text": "Research Paper: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks" + }, + { + "span_id": "span-c517fae31d25a87bf9d4001f", + "text": "Blog: Retrieval Augmented Generation (RAG)" + }, + { + "span_id": "span-040256ac3fdc275a04ed583b", + "text": "Deep Lake Vector Store" + }, + { + "span_id": "span-c22557142ac110dd5ac38414", + "text": "Chapter VI: Introduction to Langchain & LlamaIndex" + } + ] + }, + { + "chunk_id": "tai-0930b4f60884c7cb2d702e4a", + "text": "Notebook\nBuilding Application Powered by LLMs with LangChain (Section: Tutorial 1: Building LLM-Powered Applications with LangChain) Build a News Articles Summarizer (Section: Tutorial 2: Building a News Articles Summarizer) LlamaIndex Introduction (Section: LlamaIndex Introduction) Resources LangChain Documentation Useful LangChain Components: | Prompts | Output Parsers | Retrievers | Document Loaders | Text Splitters | Indexes | Embeddings models | Vector Stores | Agents | Chain | Tool | Memory | Callbacks Useful LangChain Agents: Zero-shot ReAct | Structured Input ReAct | OpenAI Functions Agent | Self-Ask with Search Agent | ReAct Document Store Agent | Plan-and-execute agents The output for the Building LLM-Powered Applications with LangChain section is based on The One Page Linux Manual: A summary of useful Linux commands Building applications with LLMs through composability A Complete Guide to LangChain: Building Powerful Applications with LLMs LlamaIndex Index Guide LlamaIndex: How to use Index correctly Defining a Custom Query Engine Working Example of Implementing Routers LlamaIndex documentation Financial Document Analysis with LlamaIndex LlamaIndex: Adding Personal Data to LLMs Llamahub\n\nChapter VII: Prompting with LangChain", + "heading": "Notebook", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-55b69ed14f5cfba843bb0c1e", + "text": "Building Application Powered by LLMs with LangChain (Section: Tutorial 1: Building LLM-Powered Applications with LangChain)" + }, + { + "span_id": "span-28955ad49a9f34045576d61f", + "text": "Build a News Articles Summarizer (Section: Tutorial 2: Building a News Articles Summarizer)" + }, + { + "span_id": "span-e29bc4f8f588799db20777a7", + "text": "LlamaIndex Introduction (Section: LlamaIndex Introduction)" + }, + { + "span_id": "span-755aa793469214d31e9c95fd", + "text": "LangChain Documentation" + }, + { + "span_id": "span-0c6b1cb694e10cf2ab84c098", + "text": "Useful LangChain Components: | Prompts | Output Parsers | Retrievers | Document Loaders | Text Splitters | Indexes | Embeddings models | Vector Stores | Agents | Chain | Tool | Memory | Callbacks" + }, + { + "span_id": "span-9fd5524bb470c92feab7a008", + "text": "Useful LangChain Agents: Zero-shot ReAct | Structured Input ReAct | OpenAI Functions Agent | Self-Ask with Search Agent | ReAct Document Store Agent | Plan-and-execute agents" + }, + { + "span_id": "span-6e00931c8871a9ce8a00c03a", + "text": "The output for the Building LLM-Powered Applications with LangChain section is based on The One Page Linux Manual: A summary of useful Linux commands" + }, + { + "span_id": "span-1836fc66d29d8fb481186219", + "text": "Building applications with LLMs through composability" + }, + { + "span_id": "span-64d3cebafb0fa0cec93bbc3a", + "text": "A Complete Guide to LangChain: Building Powerful Applications with LLMs" + }, + { + "span_id": "span-269cb744ca0c93c38d05c568", + "text": "LlamaIndex Index Guide" + }, + { + "span_id": "span-74ffe062feb9cf391fb1f098", + "text": "LlamaIndex: How to use Index correctly" + }, + { + "span_id": "span-16f4ff3b67fd01275e261beb", + "text": "Defining a Custom Query Engine" + }, + { + "span_id": "span-428c14c2d721a6164b8de0ac", + "text": "Working Example of Implementing Routers" + }, + { + "span_id": "span-cf2d34edc3abd796f264b31e", + "text": "LlamaIndex documentation" + }, + { + "span_id": "span-803dd6bd6e89cc8e4ca693fc", + "text": "Financial Document Analysis with LlamaIndex" + }, + { + "span_id": "span-762130ab4262217fa17ca4db", + "text": "LlamaIndex: Adding Personal Data to LLMs" + }, + { + "span_id": "span-0d84073f50f50fe252deb4a8", + "text": "Chapter VII: Prompting with LangChain" + } + ] + }, + { + "chunk_id": "tai-cb55fedb86777b63a85990bd", + "text": "Notebook\nUsing Prompt Templates (Section: What are LangChain Prompt Templates) Getting the Best of Few Shot Prompts & Example Selectors (Section: Few Shot Prompts and Example Selectors) Chains and Why They Are Used Notebook (Section: What are LangChain Chains) Managing Outputs with Output Parsers (Section: Tutorial 1: Managing Outputs with Output Parsers) Improving Our News Articles Summarizer (Section: Tutorial 2: Improving Our News Articles Summarizer) Creating Knowledge Graphs from Textual Data Unveiling Hidden Connections (Section: Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections) Resources Building a Knowledge Base from Texts: a Full Practical Example GitHub: langchain Knowledge Graph Visualization: NetworkX library Knowledge Graph Visualization: Pyvis library\n\nChapter VIII: Indexes, Retrievers, and Data Preparation\n\nNotebook\nWhat are Text Splitters and Why They are Useful (Section: Text Splitters) Create a YouTube Video Summarizer Using Whisper and LangChain (Section: Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain) Guarding Against Undesirable Outputs With the Self-Critique Chain (Section: Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain) Guarding Against Undesirable Outputs With the Self-Critique Chain Example (Section: Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot)", + "heading": "Notebook", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-63a6d9c4da5c1774583caf52", + "text": "Using Prompt Templates (Section: What are LangChain Prompt Templates)" + }, + { + "span_id": "span-809c2dff080c66e0535ecb30", + "text": "Getting the Best of Few Shot Prompts & Example Selectors (Section: Few Shot Prompts and Example Selectors)" + }, + { + "span_id": "span-479cd23c7dd6ff0a2f5593d7", + "text": "Chains and Why They Are Used Notebook (Section: What are LangChain Chains)" + }, + { + "span_id": "span-053cd378af4bff1f075d6921", + "text": "Managing Outputs with Output Parsers (Section: Tutorial 1: Managing Outputs with Output Parsers)" + }, + { + "span_id": "span-ef33b10b9584a291029e149d", + "text": "Improving Our News Articles Summarizer (Section: Tutorial 2: Improving Our News Articles Summarizer)" + }, + { + "span_id": "span-e7c766c97fb997d590ff01c4", + "text": "Creating Knowledge Graphs from Textual Data Unveiling Hidden Connections (Section: Tutorial 3: Creating Knowledge Graphs from Textual Data: Finding Hidden Connections)" + }, + { + "span_id": "span-bbf64c18760d546abb97bf6c", + "text": "Building a Knowledge Base from Texts: a Full Practical Example" + }, + { + "span_id": "span-c626f4eec13a4dcc8532b24d", + "text": "GitHub: langchain" + }, + { + "span_id": "span-205a5f26f4befe2a38fa544b", + "text": "Knowledge Graph Visualization: NetworkX library" + }, + { + "span_id": "span-e8b746b213b0205d3f7b518a", + "text": "Knowledge Graph Visualization: Pyvis library" + }, + { + "span_id": "span-169773db2e4b2f293ec32d51", + "text": "Chapter VIII: Indexes, Retrievers, and Data Preparation" + }, + { + "span_id": "span-200425c7938db782262ab9a9", + "text": "What are Text Splitters and Why They are Useful (Section: Text Splitters)" + }, + { + "span_id": "span-51eb5eb77ecbdfae0bdf2e8d", + "text": "Create a YouTube Video Summarizer Using Whisper and LangChain (Section: Tutorial 2: A YouTube Video Summarizer Using Whisper and LangChain)" + }, + { + "span_id": "span-2848d305ac47c20c779765ff", + "text": "Guarding Against Undesirable Outputs With the Self-Critique Chain (Section: Tutorial 4: Preventing Undesirable Outputs with the Self-Critique Chain)" + }, + { + "span_id": "span-5f47aaa76df0972d6040ecc5", + "text": "Guarding Against Undesirable Outputs With the Self-Critique Chain Example (Section: Tutorial 5: Preventing Undesirable Outputs from a Customer Service Chatbot)" + } + ] + }, + { + "chunk_id": "tai-c8da37b2acc9e9e67db8682d", + "text": "Book File\nBook File Requirements Sample PDF Used for Customizing Text Splitters Example: The One Page Linux Manual Tokens and APIs & Packages Ffmpeg Open AI API Keys Eleven API Keys Activeloop Token Resources Tutorial: How to install ffmpeg Documentation: Text wrapping and filling Code: idontcalculate/langchain GitHub Repo: JarvisBase LangChain Docs: Split by character | Split code | Recursively split by character | Summarization Open AI Whisper Acitveloop Docs: Deep Lake Vector Store in LangChain Documentation: Self-critique chain with constitutional AI\n\nChapter IX: Advanced RAG", + "heading": "Book File", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-da8545310be7fc465301b26f", + "text": "Book File Requirements" + }, + { + "span_id": "span-f09398c1926f19afdfa74225", + "text": "Sample PDF Used for Customizing Text Splitters Example: The One Page Linux Manual" + }, + { + "span_id": "span-3a67f348ce37198ffa75819f", + "text": "Tokens and APIs & Packages" + }, + { + "span_id": "span-dc50aa5acca369b5ddd58024", + "text": "Open AI API Keys" + }, + { + "span_id": "span-2d558843198d0f5dc8fb9e01", + "text": "Eleven API Keys" + }, + { + "span_id": "span-c3c638cac8bd30cf32c99b23", + "text": "Activeloop Token" + }, + { + "span_id": "span-74f28c2869f12df67016c8c6", + "text": "Tutorial: How to install ffmpeg" + }, + { + "span_id": "span-64ebcbd277082dd3f17105d5", + "text": "Documentation: Text wrapping and filling" + }, + { + "span_id": "span-1f35f631e2012071bff39b5d", + "text": "Code: idontcalculate/langchain" + }, + { + "span_id": "span-2eb07c196c044e53200429d8", + "text": "GitHub Repo: JarvisBase" + }, + { + "span_id": "span-61f1df800d4ec4d1952900bf", + "text": "LangChain Docs: Split by character | Split code | Recursively split by character | Summarization" + }, + { + "span_id": "span-4f9536f8e1c6fb208ae1ccfa", + "text": "Open AI Whisper" + }, + { + "span_id": "span-6a8eb687730a156bbe8ee9be", + "text": "Acitveloop Docs: Deep Lake Vector Store in LangChain" + }, + { + "span_id": "span-b78d0b741ca49fd552386771", + "text": "Documentation: Self-critique chain with constitutional AI" + }, + { + "span_id": "span-129fadf50322f32fa8f4cf30", + "text": "Chapter IX: Advanced RAG" + } + ] + }, + { + "chunk_id": "tai-65ede3414fede093483060bf", + "text": "Notebook\nMasterting Advanced RAG (Section: Using a Query Engine to Answer Queries) RAG Metrics & Evaluations (Section: RAG Metrics & Evaluation) LangSmith Introduction (Section: LangChain LangSmith and LangChain Hub) Resources Source Document for the Query Engine Example: Text Data for the example (Section: Query Engine Example) Tutorial: Building an Advanced Fusion Retriever from Scratch LangChain Docs: Query construction Cohere Reranking Tutorial: Hands-on Tutorial for Implementing Small-to-Big Retrieval Colab Notebook: Cohere Rerank Endpoint Blog: Complex Query Resolution through LlamaIndex Utilizing Recursive Retrieval, Document Agents, and Sub Question Query Decomposition Blog: Improving Retrieval Performance by Fine-tuning Cohere Reranker with LlamaIndex LlamaIndex Notebook LllamaIndex Docs (Section: The Role of the Retrieval Step): Retriever Query Engine with Custom Retrievers – Simple Hybrid Search | Metadata Filtering | Cohere Rerank | Document Summary Index LllamaIndex Docs (Section: RAG Metrics): Correctness | Faithfulness | Context Relevancy | Guideline Adherence | Embedding Semantic Similari | Finetuning Embeddings | Evaluating – LlamaIndex | Retrieval Evaluation | Golden Dataset | Response Evaluation | Recursive Retriever + Query Engine Demo | Query Engine | Chat Engine Creating the Dataset openai-cookbook RAGAS GitHub repository RagEvaluatorPack Downloading a LlamaDataset from LlamaHub LangSmith Hub-examples: LangSmith cookbook The Art of LangSmith LangServe Github Repository\n\nChapter X: Agents", + "heading": "Notebook", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-174a300c2c3d03f0a6156d43", + "text": "Masterting Advanced RAG (Section: Using a Query Engine to Answer Queries)" + }, + { + "span_id": "span-c89074f1b7be788440d18870", + "text": "RAG Metrics & Evaluations (Section: RAG Metrics & Evaluation)" + }, + { + "span_id": "span-35a8754b754c23355de20fb3", + "text": "LangSmith Introduction (Section: LangChain LangSmith and LangChain Hub)" + }, + { + "span_id": "span-efc1cdc484bed83e9f71e7a2", + "text": "Source Document for the Query Engine Example: Text Data for the example (Section: Query Engine Example)" + }, + { + "span_id": "span-b737a75a58560b9d59dd14dc", + "text": "Tutorial: Building an Advanced Fusion Retriever from Scratch" + }, + { + "span_id": "span-8ad89fb42b0aba63ffa4a558", + "text": "LangChain Docs: Query construction" + }, + { + "span_id": "span-035a2a090ed44a33e107f767", + "text": "Cohere Reranking" + }, + { + "span_id": "span-408b1652fda7c12fe2dcec8c", + "text": "Tutorial: Hands-on Tutorial for Implementing Small-to-Big Retrieval" + }, + { + "span_id": "span-5366ee32cdbb183338fa99c9", + "text": "Colab Notebook: Cohere Rerank Endpoint" + }, + { + "span_id": "span-c66e3b3702a5dad61465c5b4", + "text": "Blog: Complex Query Resolution through LlamaIndex Utilizing Recursive Retrieval, Document Agents, and Sub Question Query Decomposition" + }, + { + "span_id": "span-e739cd36a34725a7b4626533", + "text": "Blog: Improving Retrieval Performance by Fine-tuning Cohere Reranker with LlamaIndex" + }, + { + "span_id": "span-21ee3549b8b61bd929248569", + "text": "LlamaIndex Notebook" + }, + { + "span_id": "span-86da9dc0757d0651e97944ae", + "text": "LllamaIndex Docs (Section: The Role of the Retrieval Step): Retriever Query Engine with Custom Retrievers – Simple Hybrid Search | Metadata Filtering | Cohere Rerank | Document Summary Index" + }, + { + "span_id": "span-98dd3a1e2f1b5dbfc5cf0983", + "text": "Creating the Dataset" + }, + { + "span_id": "span-05055c5fcb324ea42d52aaec", + "text": "openai-cookbook" + }, + { + "span_id": "span-5d0686ce97df08c529556246", + "text": "RAGAS GitHub repository" + }, + { + "span_id": "span-e6bcc63c5be4a9d2f20a159e", + "text": "RagEvaluatorPack Downloading a LlamaDataset from LlamaHub" + }, + { + "span_id": "span-8e08c8a3b187dd4ae3107743", + "text": "Hub-examples: LangSmith cookbook" + }, + { + "span_id": "span-059c63614697ecaa262038e8", + "text": "The Art of LangSmith" + }, + { + "span_id": "span-f9c501ea1730a956adbead7b", + "text": "LangServe Github Repository" + }, + { + "span_id": "span-6f3f204a111061ead99a30da", + "text": "Chapter X: Agents" + } + ] + }, + { + "chunk_id": "tai-434ae6f4349454ed4e2b26d0", + "text": "Notebook\nUsing AutoGPT with LangChain (Section: Using AutoGPT with LangChain) Using AutoGPT with LangChain Output (Section: Using AutoGPT with LangChain) Building Autonomous Agents to Create Analysis Reports (Section: Tutorial 1: Building Agents for Analysis Report Creation) Query and Summarize a DB with LlamaIndex (Section: Tutorial 2: Query and Summarize a DB with LlamaIndex) Building Agents with OpenAI Assistants (Section: Tutorial 3: Building Agents with OpenAI Assistants) Multimodal Finance + Deep Memory (Section: Tutorial 5: Multimodal Financial Document Analysis from PDFs)\n\nDataset\nDataset for the Multimodal Financial Document Analysis Example: Tesla Q3 Financial Report Preprocessed Text/Label for the Multimodal Financial Document Analysis Preprocessed Graphs for the Multimodal Financial Document Analysis Resources GitHub: Babyagi Inspired Projects GitHub: Agent Simulations GitHub: CAMEL Role-Playing Autonomous Cooperative Agents GitHub; LlamaHub LlamaIndex Docs: Data Agents | OpenAI Agent with Query Engine Tools | Multi-Document Agents Blog: OpenAI Assistants API: Walk-through and Coding a Research Assistant GitHub: HuggingFace Inference Community Colab Notebook: Assistants API OpenAI Docs: OpenAI Knowledge Retrieval Blog: Function Calling OpenAI GitHub: LangChain OpenGPTs Blog: Maximizing LangChain Efficiency: Agents and ReAct Method Review LangChain Docs: Defining Custom Tools Tutorial: Installing Poppler on Windows Tutorial: Installing Tesseract on Windows Website: AutoGPT Website: BabyAGI Research: On AutoGPT – LessWrong Website: CAMEL Research Paper: The CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society paper Research Paper: Generative Agents: Interactive Simulacra of Human Behavior Website: OpenGPTs", + "heading": "Dataset", + "index": 10, + "evidence_spans": [ + { + "span_id": "span-b6be2a8abbdc8a1014202edc", + "text": "Using AutoGPT with LangChain (Section: Using AutoGPT with LangChain)" + }, + { + "span_id": "span-9ab2914f40638fbc552f4d16", + "text": "Using AutoGPT with LangChain Output (Section: Using AutoGPT with LangChain)" + }, + { + "span_id": "span-f94fb8756926cc49a99bdb22", + "text": "Building Autonomous Agents to Create Analysis Reports (Section: Tutorial 1: Building Agents for Analysis Report Creation)" + }, + { + "span_id": "span-562ed82badde0d0b4da88973", + "text": "Query and Summarize a DB with LlamaIndex (Section: Tutorial 2: Query and Summarize a DB with LlamaIndex)" + }, + { + "span_id": "span-91afd4eaad943957e1070738", + "text": "Building Agents with OpenAI Assistants (Section: Tutorial 3: Building Agents with OpenAI Assistants)" + }, + { + "span_id": "span-05f571cb514781fb879c5626", + "text": "Multimodal Finance + Deep Memory (Section: Tutorial 5: Multimodal Financial Document Analysis from PDFs)" + }, + { + "span_id": "span-84873f11c434e94a4fece50c", + "text": "Dataset for the Multimodal Financial Document Analysis Example: Tesla Q3 Financial Report" + }, + { + "span_id": "span-8a0ecc4326c4cb304714095d", + "text": "Preprocessed Text/Label for the Multimodal Financial Document Analysis" + }, + { + "span_id": "span-75e9b6aa4b7602ce14d8b4d7", + "text": "Preprocessed Graphs for the Multimodal Financial Document Analysis" + }, + { + "span_id": "span-5876d87169fef0a897b6d03f", + "text": "GitHub: Babyagi Inspired Projects" + }, + { + "span_id": "span-54a48d3b48237a5f615a0d6b", + "text": "GitHub: Agent Simulations" + }, + { + "span_id": "span-bb025c4a21481bd5995174c4", + "text": "GitHub: CAMEL Role-Playing Autonomous Cooperative Agents" + }, + { + "span_id": "span-90dc3d5c6f3019695da15dfd", + "text": "GitHub; LlamaHub" + }, + { + "span_id": "span-40ee4e0a5f7bbd22964a9fa2", + "text": "LlamaIndex Docs: Data Agents | OpenAI Agent with Query Engine Tools | Multi-Document Agents" + }, + { + "span_id": "span-08ed46cfabad587c2ba7b6cc", + "text": "Blog: OpenAI Assistants API: Walk-through and Coding a Research Assistant" + }, + { + "span_id": "span-91ece41a89d1ab317e714e6c", + "text": "GitHub: HuggingFace Inference Community" + }, + { + "span_id": "span-9d96c66da72fdf5ffa22739e", + "text": "Colab Notebook: Assistants API" + }, + { + "span_id": "span-359f265d81ef10c309f41470", + "text": "OpenAI Docs: OpenAI Knowledge Retrieval" + }, + { + "span_id": "span-536b21e4dd06f1a0cef3e6ff", + "text": "Blog: Function Calling OpenAI" + }, + { + "span_id": "span-3ae43efbf3ae39e5e2155c73", + "text": "GitHub: LangChain OpenGPTs" + }, + { + "span_id": "span-9b7b5d06a21c00a8924cbeaf", + "text": "Blog: Maximizing LangChain Efficiency: Agents and ReAct Method Review" + }, + { + "span_id": "span-1e5dbd11880d70857865230d", + "text": "LangChain Docs: Defining Custom Tools" + }, + { + "span_id": "span-b416e9c8aaa7d418c52445c8", + "text": "Tutorial: Installing Poppler on Windows" + }, + { + "span_id": "span-eddd8871f919e5ff76d83a0f", + "text": "Tutorial: Installing Tesseract on Windows" + }, + { + "span_id": "span-68efe32669bcbba925ea04f9", + "text": "Website: AutoGPT" + }, + { + "span_id": "span-0cf44a6f93ed76df90186c2f", + "text": "Website: BabyAGI" + }, + { + "span_id": "span-332ca93c0fe07bc4ce3e2b8f", + "text": "Research: On AutoGPT – LessWrong" + }, + { + "span_id": "span-66f9afba914df39e01b8693c", + "text": "Website: CAMEL" + }, + { + "span_id": "span-ba9907136b84efb4c7568ca5", + "text": "Research Paper: The CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society paper" + }, + { + "span_id": "span-7eba2448eaf93e2b8a7d0f89", + "text": "Research Paper: Generative Agents: Interactive Simulacra of Human Behavior" + }, + { + "span_id": "span-2a278ac870bcf04755dc80df", + "text": "Website: OpenGPTs" + } + ] + }, + { + "chunk_id": "tai-0b8f74686e477acbb91ed5d4", + "text": "Chapter XI: Fine-Tuning\n\nNotebook\nFineTuning a LLM Lima CPU (Section: Tutorial 1: SFT with LoRA) FineTuning a LLM Financial Sentiment CPU (Section: Tutorial 2: Using SFT and LoRA for Financial Sentiment) Create a Dataset For Cohere Fine-Tuning (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data) Fine-Tuning Using Cohere for Medical Data (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data) Finetuning a LLM QloRA (Section: Tutorial 4/Supervised Fine-Tuning Notebook) Finetuning a Reward Model (Section: Tutorial 4/Training a Reward Model Notebook) Finetune RLHF (Section: Tutorial 4/RLHF)", + "heading": "Notebook", + "index": 11, + "evidence_spans": [ + { + "span_id": "span-10939131ae000144f1a834a9", + "text": "Chapter XI: Fine-Tuning" + }, + { + "span_id": "span-34a6718b4d1b9f45997b6c88", + "text": "FineTuning a LLM Lima CPU (Section: Tutorial 1: SFT with LoRA)" + }, + { + "span_id": "span-dbcf3e2d16eec7aa3b172cc7", + "text": "FineTuning a LLM Financial Sentiment CPU (Section: Tutorial 2: Using SFT and LoRA for Financial Sentiment)" + }, + { + "span_id": "span-e046ec29c855cf8431e80002", + "text": "Create a Dataset For Cohere Fine-Tuning (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data)" + }, + { + "span_id": "span-54ccc60b73fa953158495da4", + "text": "Fine-Tuning Using Cohere for Medical Data (Section: Tutorial 3: Fine-Tuning a Cohere LLM with Medical Data)" + }, + { + "span_id": "span-8fe29d186f9acfb46bd8ce8a", + "text": "Finetuning a LLM QloRA (Section: Tutorial 4/Supervised Fine-Tuning Notebook)" + }, + { + "span_id": "span-13151e5774c27b77b76578ac", + "text": "Finetuning a Reward Model (Section: Tutorial 4/Training a Reward Model Notebook)" + }, + { + "span_id": "span-f985b30be87c14b3387eba96", + "text": "Finetune RLHF (Section: Tutorial 4/RLHF)" + } + ] + }, + { + "chunk_id": "tai-62a9dffc211047f0e5b90c3d", + "text": "Book Model Checkpoints, Requirements, Datasets, W&B Reports\nOPT fine-tuned LIMA checkpoint on CPU (Section: Practical Example: SFT with LoRA) OPT Fine-tuned finGPT with CPU (Section: Using SFT for Financial Sentiment) The Merged Model Checkpoint (2GB) (Section: Supervised Fine-Tuning Notebook) Requirements (Section: Supervised Fine-Tuning Notebook) The Reward Model Checkpoint (Step 1000 – 2GB) (Section: Training a Reward Model Notebook) Requirements (Section: Training a Reward Model Notebook) The Merged RL Model Checkpoint (2GB) (Section: RLHF) Requirements (Section: RLHF) BC5CDR Dataset in JSON format (Section: Fine-Tuning a Cohere LLM with Medical Data) Preprocessed Dataset (Section: Fine-Tuning a Cohere LLM with Medical Data) Complete Dataset (Section: Supervised Fine-Tuning Notebook) OpenOrca Dataset (Section: Supervised Fine-Tuning Notebook & Section: RLHF) “helpfulness/harmless”: (hh) by Anthropic (Section: Training a Reward Model Notebook) OPT Fine-tuned LIMA CPU (Section: Practical Example: SFT with LoRA) Weights & Bias Report (Section: Supervised Fine-Tuning Notebook) Weights & Bias Report (Section: Training a Reward Model Notebook) Weights and Biases report (Section: RLHF) Resources Research Paper: Low-Rank Adaptation (LoRA) Research Paper: QLoRA: An Efficient Variant of LoRA Open-source Resources for LoRA: PEFT Library | Lit-GPT Cohere Docs: Fine-tuning an Embedding Model for Classification Research Paper: Reinforcement Learning from Human Feedback Research Paper: LIMA: Less Is More for Alignment Research Paper: Direct Preference Optimization (DPO) Research Paper: Google DeepMind’s Reinforced Self-Training (ReST) Research Paper: Reinforcement Learning from AI Feedback (RLAIF)", + "heading": "Book Model Checkpoints, Requirements, Datasets, W&B Reports", + "index": 12, + "evidence_spans": [ + { + "span_id": "span-4077b1b283dc92d6e857d8b0", + "text": "OPT fine-tuned LIMA checkpoint on CPU (Section: Practical Example: SFT with LoRA)" + }, + { + "span_id": "span-7bdf2ed21b4e0f099d52f483", + "text": "OPT Fine-tuned finGPT with CPU (Section: Using SFT for Financial Sentiment)" + }, + { + "span_id": "span-4dacd438d87f005ef5594b07", + "text": "The Merged Model Checkpoint (2GB) (Section: Supervised Fine-Tuning Notebook)" + }, + { + "span_id": "span-8afef687289216f3b3eb96f6", + "text": "Requirements (Section: Supervised Fine-Tuning Notebook)" + }, + { + "span_id": "span-f24eca5e5a3dd342e0ae9a16", + "text": "The Reward Model Checkpoint (Step 1000 – 2GB) (Section: Training a Reward Model Notebook)" + }, + { + "span_id": "span-97613a342b4b9dfd7def7a64", + "text": "Requirements (Section: Training a Reward Model Notebook)" + }, + { + "span_id": "span-469f4e3b8d68aa8a6ffd395f", + "text": "The Merged RL Model Checkpoint (2GB) (Section: RLHF)" + }, + { + "span_id": "span-5d8318c9838786d10c3337b2", + "text": "Requirements (Section: RLHF)" + }, + { + "span_id": "span-6a67e47cc7f57db12fa989a1", + "text": "BC5CDR Dataset in JSON format (Section: Fine-Tuning a Cohere LLM with Medical Data)" + }, + { + "span_id": "span-8e1b852bc55b979e3b8118b2", + "text": "Preprocessed Dataset (Section: Fine-Tuning a Cohere LLM with Medical Data)" + }, + { + "span_id": "span-9de1cbb965841a396a88de24", + "text": "Complete Dataset (Section: Supervised Fine-Tuning Notebook)" + }, + { + "span_id": "span-15eb4377966a0e5eed2f0d77", + "text": "OpenOrca Dataset (Section: Supervised Fine-Tuning Notebook & Section: RLHF)" + }, + { + "span_id": "span-bf6515325f69423a9d22fd26", + "text": "“helpfulness/harmless”: (hh) by Anthropic (Section: Training a Reward Model Notebook)" + }, + { + "span_id": "span-73a8201fa5ff0e7eeac37989", + "text": "OPT Fine-tuned LIMA CPU (Section: Practical Example: SFT with LoRA)" + }, + { + "span_id": "span-a0e1a03a751824ebb1165ab1", + "text": "Weights & Bias Report (Section: Supervised Fine-Tuning Notebook)" + }, + { + "span_id": "span-3a879f832d687177e0e4f0e8", + "text": "Weights & Bias Report (Section: Training a Reward Model Notebook)" + }, + { + "span_id": "span-6feede9028668a5032baa1dd", + "text": "Weights and Biases report (Section: RLHF)" + }, + { + "span_id": "span-575cc8110fa2c4935e8d3e40", + "text": "Research Paper: Low-Rank Adaptation (LoRA)" + }, + { + "span_id": "span-8e340eb0dcf181c1abd5381d", + "text": "Research Paper: QLoRA: An Efficient Variant of LoRA" + }, + { + "span_id": "span-7e78a2571641bbfad59919c2", + "text": "Open-source Resources for LoRA: PEFT Library | Lit-GPT" + }, + { + "span_id": "span-c7b1e3faf71df63d430cb41f", + "text": "Cohere Docs: Fine-tuning an Embedding Model for Classification" + }, + { + "span_id": "span-a111a8f14bf4b9cdab81bfab", + "text": "Research Paper: Reinforcement Learning from Human Feedback" + }, + { + "span_id": "span-e67506ea24a42855e26059fe", + "text": "Research Paper: LIMA: Less Is More for Alignment" + }, + { + "span_id": "span-566049e82dc7c2dd34cbb234", + "text": "Research Paper: Direct Preference Optimization (DPO)" + }, + { + "span_id": "span-3cf60a6a71624b00c1cdfc2d", + "text": "Research Paper: Google DeepMind’s Reinforced Self-Training (ReST)" + }, + { + "span_id": "span-2dfe33de3f6cc0c84e1e249a", + "text": "Research Paper: Reinforcement Learning from AI Feedback (RLAIF)" + } + ] + }, + { + "chunk_id": "tai-f16dc6ac28da4c85c5bb7f1d", + "text": "Chapter XII: Deployment\n\nNotebook\nBenchmark Inference (Section: Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP)) Resources Research Paper: Model Compression Research Paper: Distilling the Knowledge in a Neural Network Research Paper: A Survey of Quantization Methods for Efficient Neural Network Inference Research Paper: Sparsity in Deep Learning GitHub: Hugging Face Optimum GitHub: Intel Neural Compressor Research Paper: LLM.int8(): 8bit Matrix Multiplication for Transformers at Scale Research Paper: GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers Research Paper: AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration Research Paper: Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures Reasearch Paper: A Simple and Effective Pruning Approach for Large Language Models Reasearch Paper: Structured Pruning of Deep Convolutional Neural Networks Research Paper: The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks A complete list of tasks supported with Simple Quantization (Using CLI) The Docker Image under Latitude: Llama 2 API Inference\n\nConclusion\nNo Notebooks.\n\nFurther Reading and Courses\nTowards AI Open Source AI-Tutor HuggingFace: AI Tutor Chatbot\n\nPrevious Courses\nTraining & Fine-Tuning LLMs for Production LangChain & Vector Databases in Production", + "heading": "Notebook", + "index": 13, + "evidence_spans": [ + { + "span_id": "span-b253ba28c32594d18fdfb9f6", + "text": "Chapter XII: Deployment" + }, + { + "span_id": "span-2b16ed354db88ae671f56bf3", + "text": "Benchmark Inference (Section: Tutorial: Deploying a Quantized LLM on a CPU on Google Cloud Platform (GCP))" + }, + { + "span_id": "span-c3a42ba47447af858a17077f", + "text": "Research Paper: Model Compression" + }, + { + "span_id": "span-f95a1fad7835ad3ed086843b", + "text": "Research Paper: Distilling the Knowledge in a Neural Network" + }, + { + "span_id": "span-73d2721e5d0300f9742b2391", + "text": "Research Paper: A Survey of Quantization Methods for Efficient Neural Network Inference" + }, + { + "span_id": "span-7c19fc70cf09a788d20da9a8", + "text": "Research Paper: Sparsity in Deep Learning" + }, + { + "span_id": "span-afef86e8f65ca1375451ccc1", + "text": "GitHub: Hugging Face Optimum" + }, + { + "span_id": "span-54a6548549149075cd6fa0ca", + "text": "GitHub: Intel Neural Compressor" + }, + { + "span_id": "span-52f44ab671be04a4f1fe77e2", + "text": "Research Paper: LLM.int8(): 8bit Matrix Multiplication for Transformers at Scale" + }, + { + "span_id": "span-6ce1b35f5823ae6c5ddd5026", + "text": "Research Paper: GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers" + }, + { + "span_id": "span-a3018887fe04bcca5060caf5", + "text": "Research Paper: AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration" + }, + { + "span_id": "span-9692222c1569e56a3cfe96e0", + "text": "Research Paper: Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures" + }, + { + "span_id": "span-a8970173e6b2d79506cb0f37", + "text": "Reasearch Paper: A Simple and Effective Pruning Approach for Large Language Models" + }, + { + "span_id": "span-01c9d80c4e76f9b8cbfcc343", + "text": "Reasearch Paper: Structured Pruning of Deep Convolutional Neural Networks" + }, + { + "span_id": "span-8bafe1a2f4ac9eff15376841", + "text": "Research Paper: The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks" + }, + { + "span_id": "span-2c75280db26bf634ca02611a", + "text": "A complete list of tasks supported with Simple Quantization (Using CLI)" + }, + { + "span_id": "span-cf05050905fb6733711b57c6", + "text": "The Docker Image under Latitude: Llama 2 API Inference" + }, + { + "span_id": "span-a0a4bc248f99528d2ca691b4", + "text": "No Notebooks." + }, + { + "span_id": "span-103ed9b45eb9ff402eebbc74", + "text": "Towards AI Open Source AI-Tutor" + }, + { + "span_id": "span-c51640a96b042906702d4248", + "text": "HuggingFace: AI Tutor Chatbot" + }, + { + "span_id": "span-1ec95f3c57168641f9133a23", + "text": "Training & Fine-Tuning LLMs for Production" + }, + { + "span_id": "span-34d2e1091e69417025fcd587", + "text": "LangChain & Vector Databases in Production" + } + ] + }, + { + "chunk_id": "tai-7cab537c1f5125cf6d0b78cd", + "text": "Free Resources\nTowards AI Newsletter What’s AI YouTube Channel Towards AI Academy Note: This webpage has been updated to follow the order and structure of the second edition. All resources from the first edition are still available but may have been rearranged to match the new sequence. Additionally, you’ll find new links and resources exclusive to the second edition—an added benefit for readers with the first edition.", + "heading": "Free Resources", + "index": 14, + "evidence_spans": [ + { + "span_id": "span-e3534abbd2dcbe07f14a7843", + "text": "Towards AI Newsletter" + }, + { + "span_id": "span-d7071fe7cfe95ab09110d1fa", + "text": "What’s AI YouTube Channel" + }, + { + "span_id": "span-d61acec17e499adae8562679", + "text": "Towards AI Academy" + }, + { + "span_id": "span-7581082fa390a0e627fcb02d", + "text": "Note: This webpage has been updated to follow the order and structure of the second edition." + }, + { + "span_id": "span-5e10f6367f98d854009e968d", + "text": "All resources from the first edition are still available but may have been rearranged to match the new sequence." + }, + { + "span_id": "span-82eba26d6d1d6f3d2c523703", + "text": "Additionally, you’ll find new links and resources exclusive to the second edition—an added benefit for readers with the first edition." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "a3123e337e873bf32ebe9d52dfaf654506f20a8ec207ffc060f4eb7dddbd3161", + "content_hash": "a3123e337e873bf32ebe9d52dfaf654506f20a8ec207ffc060f4eb7dddbd3161", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/book/" + ], + "evidence_hash": "15d5d755780eb587e7e6c747ea1aba58d7b81cd75f90928e1d6255b3d96cbe6b" + }, + { + "discovered_url": "https://towardsai.com/academy/bundles/get-it-all/", + "url": "https://towardsai.com/academy/bundles/get-it-all/", + "canonical_url": "https://towardsai.com/academy/bundles/get-it-all/", + "host": "towardsai.com", + "path": "/academy/bundles/get-it-all", + "kind": "bundle", + "offer_id": "get-it-all", + "entity_id": "offer:get-it-all", + "title": "Get It All: Every Course, One Bundle | Towards AI Academy", + "review_title": "Get It All", + "reviewed_summary": "Broadest Academy course bundle.", + "meta_description": "All five Towards AI courses plus the Building LLMs for Production e-book in one bundle: Python, LLM fundamentals, full stack AI engineering, agents, and AI for work.", + "headings": [ + "Get it all. Novice to expert", + "Every product, one purchase", + "Beginner Python for AI Engineering", + "10-Hour LLM Fundamentals", + "Building LLMs for Production", + "Full Stack AI Engineering", + "Agent Engineering", + "Master AI for Work", + "30-day money-back guarantee", + "Frequently asked questions", + "The complete path from first script to shipped AI" + ], + "text": "Course bundle · for everyone\n\nGet it all. Novice to expert\nThe complete curriculum for anyone: learn Python, get fluent with LLMs, and become proficient at building with them. Every course we make, plus the book, in one bundle. 6 products in the bundle $1,625 combined price bought separately 5.0 average course rating All access one bundle · lifetime access 10-Hour LLM Fundamentals ($199) Beginner Python for AI Engineering ($149) Full Stack AI Engineering ($349) Agent Engineering ($499) Master AI for Work ($399) Building LLMs for Production, the e-book ($29.99) Become an AI expert Bundle price shown at checkout in your currency · 30-day money-back guarantee 01 Bundle includes\n\nEvery product, one purchase\nAll six products in the bundle, from first line of Python to deployed multi-agent systems. $149 · rated 5.0\n\nBeginner Python for AI Engineering\nA Python course for the LLM era. Build a foundation for developing your own LLM apps: API integration, open-source models, and fundamental training and testing skills. $199 · rated 5.0\n\n10-Hour LLM Fundamentals\nFive two-hour video sessions that take you from a basic prompt to a full production rollout, with the essential skills for production-ready applications. $29.99 · the e-book\n\nBuilding LLMs for Production\nTechniques that scale to enterprise workflows, helping independent developers and small teams with limited resources create sellable AI products. $349 · rated 5.0 (43 reviews)\n\nFull Stack AI Engineering\nBuild a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment. $499 · rated 5.0 (26 reviews)\n\nAgent Engineering\nDesign, build, evaluate, and deploy production-grade AI agents. Two deployed systems and the engineering judgment to architect any agentic product. $399 · rated 5.0\n\nMaster AI for Work\nFor anyone who wants AI productivity and capability gains across real work tasks, and for leaders bringing AI into their company and team. No code required. Risk-free learning\n\n30-day money-back guarantee\nWe curate every course with a practical-first approach, but learning is personal. If a course isn't the right fit, tell us within 30 days and we'll make it right with a full refund. Start risk-free 02 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries. Six products · one bundle\n\nThe complete path from first script to shipped AI\nBecome an AI expert Compare the courses", + "links": [ + { + "text": "Become an AI expert", + "url": "https://academy.towardsai.net/cart/add_product/3287337?price_id=4184611" + }, + { + "text": "", + "url": "https://towardsai.com/academy/python-for-ai-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/book/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/agent-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/ai-for-work/" + }, + { + "text": "Start risk-free", + "url": "https://academy.towardsai.net/cart/add_product/3287337?price_id=4184611" + }, + { + "text": "Affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + }, + { + "text": "Compare the courses", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-700f943d935474284ed48ae5", + "text": "Course bundle · for everyone\n\nGet it all. Novice to expert\nThe complete curriculum for anyone: learn Python, get fluent with LLMs, and become proficient at building with them. Every course we make, plus the book, in one bundle. 6 products in the bundle $1,625 combined price bought separately 5.0 average course rating All access one bundle · lifetime access 10-Hour LLM Fundamentals ($199) Beginner Python for AI Engineering ($149) Full Stack AI Engineering ($349) Agent Engineering ($499) Master AI for Work ($399) Building LLMs for Production, the e-book ($29.99) Become an AI expert Bundle price shown at checkout in your currency · 30-day money-back guarantee 01 Bundle includes\n\nEvery product, one purchase\nAll six products in the bundle, from first line of Python to deployed multi-agent systems. $149 · rated 5.0\n\nBeginner Python for AI Engineering\nA Python course for the LLM era. Build a foundation for developing your own LLM apps: API integration, open-source models, and fundamental training and testing skills. $199 · rated 5.0\n\n10-Hour LLM Fundamentals\nFive two-hour video sessions that take you from a basic prompt to a full production rollout, with the essential skills for production-ready applications. $29.99 · the e-book\n\nBuilding LLMs for Production\nTechniques that scale to enterprise workflows, helping independent developers and small teams with limited resources create sellable AI products. $349 · rated 5.0 (43 reviews)\n\nFull Stack AI Engineering\nBuild a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment. $499 · rated 5.0 (26 reviews)", + "heading": "Get it all. Novice to expert", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-1affcbbea077f7efb7503eb8", + "text": "Course bundle · for everyone" + }, + { + "span_id": "span-b9181f94c3abd7f4520fc191", + "text": "The complete curriculum for anyone: learn Python, get fluent with LLMs, and become proficient at building with them." + }, + { + "span_id": "span-80c84c4f596640bebe7350a4", + "text": "Every course we make, plus the book, in one bundle." + }, + { + "span_id": "span-d1f6b0e09580757488d12394", + "text": "6 products in the bundle" + }, + { + "span_id": "span-d12b943ff6592c3a65683ae0", + "text": "$1,625 combined price bought separately" + }, + { + "span_id": "span-13fb1dbcbd1b83591641ac4e", + "text": "5.0 average course rating" + }, + { + "span_id": "span-8e89183102c56e65917de895", + "text": "All access one bundle · lifetime access" + }, + { + "span_id": "span-ea98bb9c3c840f27f886d58b", + "text": "10-Hour LLM Fundamentals ($199) Beginner Python for AI Engineering ($149) Full Stack AI Engineering ($349) Agent Engineering ($499) Master AI for Work ($399) Building LLMs for Production, the e-book ($29.99)" + }, + { + "span_id": "span-3e740fcb9cddcbd4500a7248", + "text": "Become an AI expert Bundle price shown at checkout in your currency · 30-day money-back guarantee" + }, + { + "span_id": "span-ad65d002e5441ef7d1ca0b1c", + "text": "01 Bundle includes" + }, + { + "span_id": "span-faa4ed32c1ccbd1c470b07f3", + "text": "All six products in the bundle, from first line of Python to deployed multi-agent systems." + }, + { + "span_id": "span-2e7a018187b87f6b6b6649f6", + "text": "$149 · rated 5.0" + }, + { + "span_id": "span-ff5f37a2d63dbdadc1218bb8", + "text": "A Python course for the LLM era." + }, + { + "span_id": "span-2b1973537fdb9839640b579d", + "text": "Build a foundation for developing your own LLM apps: API integration, open-source models, and fundamental training and testing skills." + }, + { + "span_id": "span-40bb5a34d0106904982174e4", + "text": "$199 · rated 5.0" + }, + { + "span_id": "span-e6e826aa23164b4b57d3bda0", + "text": "10-Hour LLM Fundamentals" + }, + { + "span_id": "span-65273739d0ef90c4845c3f32", + "text": "Five two-hour video sessions that take you from a basic prompt to a full production rollout, with the essential skills for production-ready applications." + }, + { + "span_id": "span-34f330bfb5edd32723e5fcf1", + "text": "$29.99 · the e-book" + }, + { + "span_id": "span-da23c2990bf11f14b566301f", + "text": "Techniques that scale to enterprise workflows, helping independent developers and small teams with limited resources create sellable AI products." + }, + { + "span_id": "span-66f73fc2c0c15542c310cb82", + "text": "$349 · rated 5.0 (43 reviews)" + }, + { + "span_id": "span-7b4f866a9a69c8611113df3a", + "text": "Build a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment." + }, + { + "span_id": "span-4615578c9e008506c3eb5c3a", + "text": "$499 · rated 5.0 (26 reviews)" + } + ] + }, + { + "chunk_id": "tai-daa5e9909f731f10ca4a08c3", + "text": "Agent Engineering\nDesign, build, evaluate, and deploy production-grade AI agents. Two deployed systems and the engineering judgment to architect any agentic product. $399 · rated 5.0\n\nMaster AI for Work\nFor anyone who wants AI productivity and capability gains across real work tasks, and for leaders bringing AI into their company and team. No code required. Risk-free learning\n\n30-day money-back guarantee\nWe curate every course with a practical-first approach, but learning is personal. If a course isn't the right fit, tell us within 30 days and we'll make it right with a full refund. Start risk-free 02 Questions", + "heading": "30-day money-back guarantee", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-c6024aaf1bb214d814c03ea1", + "text": "Design, build, evaluate, and deploy production-grade AI agents." + }, + { + "span_id": "span-335ca223c9292cd58d028fa1", + "text": "Two deployed systems and the engineering judgment to architect any agentic product." + }, + { + "span_id": "span-00a750a9e033c3b625b72150", + "text": "$399 · rated 5.0" + }, + { + "span_id": "span-59aaf5382e8c88998b24a65b", + "text": "For anyone who wants AI productivity and capability gains across real work tasks, and for leaders bringing AI into their company and team." + }, + { + "span_id": "span-f7a9ab238fa4d9665cf44f1b", + "text": "No code required." + }, + { + "span_id": "span-4c68894874f97bb6b017c9af", + "text": "Risk-free learning" + }, + { + "span_id": "span-b985df7a2fb6cfde53451d35", + "text": "30-day money-back guarantee" + }, + { + "span_id": "span-8ede61b0511aaeae0c4452ce", + "text": "We curate every course with a practical-first approach, but learning is personal." + }, + { + "span_id": "span-f7362aac8266b5f3f9b1bab1", + "text": "If a course isn't the right fit, tell us within 30 days and we'll make it right with a full refund." + }, + { + "span_id": "span-84d58beee42b9cef24627700", + "text": "Start risk-free" + }, + { + "span_id": "span-3719472998619d9c0f12bd47", + "text": "02 Questions" + } + ] + }, + { + "chunk_id": "tai-1d7392ac031d84b6138ff663", + "text": "Frequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries. Six products · one bundle\n\nThe complete path from first script to shipped AI\nBecome an AI expert Compare the courses", + "heading": "Frequently asked questions", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-3320da5c09e96001fe127a9c", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-0fc5e13edf0099397a51445b", + "text": "The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves." + }, + { + "span_id": "span-906d4dfbd7f2b70372f4b0b1", + "text": "You're getting a living resource designed to keep you ahead, not a static course." + }, + { + "span_id": "span-a3f88c25262b7201de9703e9", + "text": "Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more." + }, + { + "span_id": "span-9b496d5191e9d276ff265cff", + "text": "Email Louis-François at louis@towardsai.net for a follow-up." + }, + { + "span_id": "span-5eb63e85f2cf5bee00f23070", + "text": "For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book." + }, + { + "span_id": "span-4537980945e85453d5e53eb3", + "text": "We also run an affiliate program for people with an audience." + }, + { + "span_id": "span-c824a10fed3c2bf1499822fd", + "text": "See the Affiliate page or email louis@towardsai.net with proof of referral." + }, + { + "span_id": "span-c4de00030029f4537cd1cacc", + "text": "We offer course bundles and custom training solutions tailored for companies." + }, + { + "span_id": "span-ba469bb6bc7f424827c25e72", + "text": "Reach out to Louis at louis@towardsai.net ." + }, + { + "span_id": "span-6a4833d6ddd4de0d7293e005", + "text": "If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days." + }, + { + "span_id": "span-260da6ef3b7156bd15bdb879", + "text": "No hassle, no worries." + }, + { + "span_id": "span-b0c1852f8c828b6f1232874b", + "text": "Six products · one bundle" + }, + { + "span_id": "span-264de58d4c1d415328951ed2", + "text": "Become an AI expert Compare the courses" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "42c02a3d89e3a0259695eb7af7865b8c029be38f1a2966d98a8097c00ce94d57", + "content_hash": "42c02a3d89e3a0259695eb7af7865b8c029be38f1a2966d98a8097c00ce94d57", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/bundles/get-it-all/" + ], + "evidence_hash": "a9521e2e801f7001462684637fc365aefc549dcd8a80f682d46e01e3ef96aee1" + }, + { + "discovered_url": "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/", + "url": "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/", + "canonical_url": "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/", + "host": "towardsai.com", + "path": "/academy/bundles/from-coding-novice-to-advanced-llm-developer", + "kind": "bundle", + "offer_id": "from-coding-novice-to-advanced-llm-developer", + "entity_id": "offer:from-coding-novice-to-advanced-llm-developer", + "title": "From Non-Coder to AI Engineer · Course Bundle | Towards AI Academy", + "review_title": "From Non-Coder to AI Engineer", + "reviewed_summary": "Bundle path from Python foundations to AI engineering.", + "meta_description": "Go from zero to certified LLM developer for $599: Beginner Python, 10-Hour LLM Fundamentals, Full Stack AI Engineering, and the e-book in one bundle.", + "headings": [ + "From non-coder to AI engineer", + "Can three courses make me an LLM developer?", + "The AI skills companies need", + "Beginner Python for AI Engineering", + "10-Hour LLM Fundamentals + the e-book", + "Full Stack AI Engineering", + "Can anyone become an LLM developer?", + "The AI skills you need, without the wrong path", + "Frequently asked questions" + ], + "text": "Course bundle · no coding background needed\n\nFrom non-coder to AI engineer\nGo from zero to a certified LLM developer. Three courses and our e-book: a complete hands-on roadmap to learning Python, mastering today's LLMs, and creating scalable LLM applications. $599 bundle price, was $727 4 products, from Python to production 5.0 average course rating $727 $599 one-time · lifetime access Beginner Python for AI Engineering ($149) 10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Building LLMs for Production, the e-book ($29.99) Become an LLM product pro 30-day money-back guarantee, no questions asked 01 The method\n\nCan three courses make me an LLM developer?\nMost AI courses throw information at you. We make sure you actually learn how to think, adapt, and build like an AI engineer. 01 Python specifically for AI development. No fluff, just the essentials to start working with LLMs and building AI-driven solutions. 02 Long-term foundations. You won't master AI overnight, but you'll gain the frameworks, workflows, and best practices to build, iterate, and scale real applications. 03 Lead AI innovation at work. Experiment effectively, optimize workflows, and deploy AI-powered solutions that create real impact. 04 Stay ahead of a fast field. Learn to stay adaptable, keep learning, and apply your skills to new challenges as AI evolves. 05 Turn skills into opportunities. Land a high-impact AI role, automate your own workflows, or launch AI-powered products. 02 Bundle includes\n\nThe AI skills companies need\nMaster Python, learn production-ready AI techniques, and build scalable LLM-powered applications. Step 1 · $149 · rated 5.0\n\nBeginner Python for AI Engineering\nPython with a build-first approach: API integration, open-source models, faster coding with AI assistants, and simple Python-for-LLMs projects. Step 2 · $199 + $29.99 · rated 5.0\n\n10-Hour LLM Fundamentals + the e-book\nWork with leading AI models and learn timeless LLM principles: prompting, RAG, fine-tuning, agents, and tools, in five two-hour sessions plus the book. Step 3 · $349 · rated 5.0 (43 reviews)\n\nFull Stack AI Engineering\nProduction-ready implementation of RAG, fine-tuning, agents, and tools, finishing with an advanced MVP for an LLM product and deployment optimization. 03 Straight answers\n\nCan anyone become an LLM developer?\nYes. The best time to start was yesterday. The second-best time is today. 01 \"I don't have years to learn AI.\" You don't need years. With a structured, hands-on approach you'll go from zero to building real AI applications in months. 02 \"I'm not a software engineer.\" We break down Python, AI fundamentals, and LLMs step by step, so you can start building even if you're new to coding. 03 \"Why build when I can use AI tools?\" Off-the-shelf tools can't always do what you need. With Python and LLM development you'll create custom solutions that fit your company. 04 \"Won't my skills become outdated?\" We teach foundational AI principles that stand the test of time, plus how to adapt and stay ahead as AI evolves. 05 \"I don't know where to start.\" That's exactly why this bundle exists: a clear, structured roadmap from Python to full-scale LLM applications. $599, was $727 · 30-day guarantee\n\nThe AI skills you need, without the wrong path\nIf you've been wanting to break into AI but didn't know where to start, this is your roadmap: from absolute beginner to confidently building enterprise workflows and LLM applications. Grab your AI roadmap Compare the courses 04 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries.", + "links": [ + { + "text": "Become an LLM product pro", + "url": "https://academy.towardsai.net/cart/add_product/3262189?price_id=4157308" + }, + { + "text": "", + "url": "https://towardsai.com/academy/python-for-ai-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/book/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "Grab your AI roadmap", + "url": "https://academy.towardsai.net/cart/add_product/3262189?price_id=4157308" + }, + { + "text": "Compare the courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + } + ], + "chunks": [ + { + "chunk_id": "tai-08133c4834d878a6f76d0d17", + "text": "Course bundle · no coding background needed\n\nFrom non-coder to AI engineer\nGo from zero to a certified LLM developer. Three courses and our e-book: a complete hands-on roadmap to learning Python, mastering today's LLMs, and creating scalable LLM applications. $599 bundle price, was $727 4 products, from Python to production 5.0 average course rating $727 $599 one-time · lifetime access Beginner Python for AI Engineering ($149) 10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Building LLMs for Production, the e-book ($29.99) Become an LLM product pro 30-day money-back guarantee, no questions asked 01 The method\n\nCan three courses make me an LLM developer?\nMost AI courses throw information at you. We make sure you actually learn how to think, adapt, and build like an AI engineer. 01 Python specifically for AI development. No fluff, just the essentials to start working with LLMs and building AI-driven solutions. 02 Long-term foundations. You won't master AI overnight, but you'll gain the frameworks, workflows, and best practices to build, iterate, and scale real applications. 03 Lead AI innovation at work. Experiment effectively, optimize workflows, and deploy AI-powered solutions that create real impact. 04 Stay ahead of a fast field. Learn to stay adaptable, keep learning, and apply your skills to new challenges as AI evolves. 05 Turn skills into opportunities. Land a high-impact AI role, automate your own workflows, or launch AI-powered products. 02 Bundle includes\n\nThe AI skills companies need\nMaster Python, learn production-ready AI techniques, and build scalable LLM-powered applications. Step 1 · $149 · rated 5.0", + "heading": "Can three courses make me an LLM developer?", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-37f29a6e957be892f2e2ba9b", + "text": "Course bundle · no coding background needed" + }, + { + "span_id": "span-a532ef30658bafabcc385c17", + "text": "Go from zero to a certified LLM developer." + }, + { + "span_id": "span-39ee13328725fd19ea942ea2", + "text": "Three courses and our e-book: a complete hands-on roadmap to learning Python, mastering today's LLMs, and creating scalable LLM applications." + }, + { + "span_id": "span-c283724306b231fd6de89b6d", + "text": "$599 bundle price, was $727" + }, + { + "span_id": "span-ac39ad1f448abaff39a5acbb", + "text": "4 products, from Python to production" + }, + { + "span_id": "span-17650ec5376df142a062c979", + "text": "5.0 average course rating" + }, + { + "span_id": "span-0dfda23714ebc1f29ceba0c7", + "text": "$727 $599 one-time · lifetime access" + }, + { + "span_id": "span-3d92b2437b54956f112fe16b", + "text": "Beginner Python for AI Engineering ($149) 10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Building LLMs for Production, the e-book ($29.99)" + }, + { + "span_id": "span-5b8510bacbbb16e373ecf97f", + "text": "Become an LLM product pro 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-f2d6b86d7464f5dcf04a28a2", + "text": "01 The method" + }, + { + "span_id": "span-f8a27f95d7d8d93a02ab74cc", + "text": "Most AI courses throw information at you." + }, + { + "span_id": "span-da615b34a45bdab79f499fb0", + "text": "We make sure you actually learn how to think, adapt, and build like an AI engineer." + }, + { + "span_id": "span-c9706546f21a38519b6510cb", + "text": "01 Python specifically for AI development." + }, + { + "span_id": "span-5d36a387e7cf0ed6f350a18a", + "text": "No fluff, just the essentials to start working with LLMs and building AI-driven solutions." + }, + { + "span_id": "span-5c38b121e103b96289279d82", + "text": "02 Long-term foundations." + }, + { + "span_id": "span-b7de5f29d832d476ac22cdd7", + "text": "You won't master AI overnight, but you'll gain the frameworks, workflows, and best practices to build, iterate, and scale real applications." + }, + { + "span_id": "span-dfade403646790d96eaf9557", + "text": "03 Lead AI innovation at work." + }, + { + "span_id": "span-863fa97b60410fecab985203", + "text": "Experiment effectively, optimize workflows, and deploy AI-powered solutions that create real impact." + }, + { + "span_id": "span-face86d1bcb4aa263e078020", + "text": "04 Stay ahead of a fast field." + }, + { + "span_id": "span-7ff2f0eda84edc1a7da67fb5", + "text": "Learn to stay adaptable, keep learning, and apply your skills to new challenges as AI evolves." + }, + { + "span_id": "span-ede27b6ac962340a8fc39063", + "text": "05 Turn skills into opportunities." + }, + { + "span_id": "span-0278c13276c61ef243feb8df", + "text": "Land a high-impact AI role, automate your own workflows, or launch AI-powered products." + }, + { + "span_id": "span-e0c68976a0b50c135ba07703", + "text": "02 Bundle includes" + }, + { + "span_id": "span-495a84ea48cd8e5350b15ece", + "text": "Master Python, learn production-ready AI techniques, and build scalable LLM-powered applications." + }, + { + "span_id": "span-99042f28ba401b93d2acc72b", + "text": "Step 1 · $149 · rated 5.0" + } + ] + }, + { + "chunk_id": "tai-cb61805f61dc6f50d0e9b7eb", + "text": "Beginner Python for AI Engineering\nPython with a build-first approach: API integration, open-source models, faster coding with AI assistants, and simple Python-for-LLMs projects. Step 2 · $199 + $29.99 · rated 5.0\n\n10-Hour LLM Fundamentals + the e-book\nWork with leading AI models and learn timeless LLM principles: prompting, RAG, fine-tuning, agents, and tools, in five two-hour sessions plus the book. Step 3 · $349 · rated 5.0 (43 reviews)\n\nFull Stack AI Engineering\nProduction-ready implementation of RAG, fine-tuning, agents, and tools, finishing with an advanced MVP for an LLM product and deployment optimization. 03 Straight answers\n\nCan anyone become an LLM developer?\nYes. The best time to start was yesterday. The second-best time is today. 01 \"I don't have years to learn AI.\" You don't need years. With a structured, hands-on approach you'll go from zero to building real AI applications in months. 02 \"I'm not a software engineer.\" We break down Python, AI fundamentals, and LLMs step by step, so you can start building even if you're new to coding. 03 \"Why build when I can use AI tools?\" Off-the-shelf tools can't always do what you need. With Python and LLM development you'll create custom solutions that fit your company. 04 \"Won't my skills become outdated?\" We teach foundational AI principles that stand the test of time, plus how to adapt and stay ahead as AI evolves. 05 \"I don't know where to start.\" That's exactly why this bundle exists: a clear, structured roadmap from Python to full-scale LLM applications. $599, was $727 · 30-day guarantee", + "heading": "Can anyone become an LLM developer?", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-95d12c96e5f9f0e835879d09", + "text": "Python with a build-first approach: API integration, open-source models, faster coding with AI assistants, and simple Python-for-LLMs projects." + }, + { + "span_id": "span-046f997ac128a776aae70ea0", + "text": "Step 2 · $199 + $29.99 · rated 5.0" + }, + { + "span_id": "span-aabdaba98a48fb1af83718ef", + "text": "10-Hour LLM Fundamentals + the e-book" + }, + { + "span_id": "span-0d0b78f67feda95366d8e398", + "text": "Work with leading AI models and learn timeless LLM principles: prompting, RAG, fine-tuning, agents, and tools, in five two-hour sessions plus the book." + }, + { + "span_id": "span-f781e8d653ca339c6314818b", + "text": "Step 3 · $349 · rated 5.0 (43 reviews)" + }, + { + "span_id": "span-33d3d4b6536738319b0353f7", + "text": "Production-ready implementation of RAG, fine-tuning, agents, and tools, finishing with an advanced MVP for an LLM product and deployment optimization." + }, + { + "span_id": "span-fa36641ad373916ee6a6cb98", + "text": "03 Straight answers" + }, + { + "span_id": "span-3da7349c30cb6c379736fc58", + "text": "The best time to start was yesterday." + }, + { + "span_id": "span-1da9d1ac2e287dab3af63d02", + "text": "The second-best time is today." + }, + { + "span_id": "span-661c7e721348d05387d9b527", + "text": "01 \"I don't have years to learn AI.\" You don't need years." + }, + { + "span_id": "span-32848dcda9edc2ca4d43c9ea", + "text": "With a structured, hands-on approach you'll go from zero to building real AI applications in months." + }, + { + "span_id": "span-d138346bd6b33f38bcd8fc28", + "text": "02 \"I'm not a software engineer.\" We break down Python, AI fundamentals, and LLMs step by step, so you can start building even if you're new to coding." + }, + { + "span_id": "span-deb96989894a98cb81ee33bd", + "text": "03 \"Why build when I can use AI tools?\" Off-the-shelf tools can't always do what you need." + }, + { + "span_id": "span-4151c38827b3beb4a5c91caa", + "text": "With Python and LLM development you'll create custom solutions that fit your company." + }, + { + "span_id": "span-62b1ed7f16c76a1d40bb0fd9", + "text": "04 \"Won't my skills become outdated?\" We teach foundational AI principles that stand the test of time, plus how to adapt and stay ahead as AI evolves." + }, + { + "span_id": "span-738461b12b10cbd757a933f3", + "text": "05 \"I don't know where to start.\" That's exactly why this bundle exists: a clear, structured roadmap from Python to full-scale LLM applications." + }, + { + "span_id": "span-ce9c1e05716a4421a962b241", + "text": "$599, was $727 · 30-day guarantee" + } + ] + }, + { + "chunk_id": "tai-dd8b63eff3e78b081773e75a", + "text": "The AI skills you need, without the wrong path\nIf you've been wanting to break into AI but didn't know where to start, this is your roadmap: from absolute beginner to confidently building enterprise workflows and LLM applications. Grab your AI roadmap Compare the courses 04 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries.", + "heading": "Frequently asked questions", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-918b2e60f8db9e8fc7f08311", + "text": "If you've been wanting to break into AI but didn't know where to start, this is your roadmap: from absolute beginner to confidently building enterprise workflows and LLM applications." + }, + { + "span_id": "span-a7aa2f139bab2fe58a48ffdf", + "text": "Grab your AI roadmap Compare the courses" + }, + { + "span_id": "span-c7c332ae19ab7d687b0852f7", + "text": "04 Questions" + }, + { + "span_id": "span-c4638ba41638744991a7c138", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-5963e1a63357c970a646cf1b", + "text": "The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves." + }, + { + "span_id": "span-2c6ce30958bed84efae3ec6f", + "text": "You're getting a living resource designed to keep you ahead, not a static course." + }, + { + "span_id": "span-134e0b6608d7af4d773f39fa", + "text": "Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more." + }, + { + "span_id": "span-6459090b65704112235ed6c1", + "text": "Email Louis-François at louis@towardsai.net for a follow-up." + }, + { + "span_id": "span-46c0919aad02b8386616cc1f", + "text": "For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book." + }, + { + "span_id": "span-65314c9b4d8279e48ed9a2be", + "text": "We also run an affiliate program for people with an audience." + }, + { + "span_id": "span-c194e4de0fc640c0587ea063", + "text": "See the Affiliate page or email louis@towardsai.net with proof of referral." + }, + { + "span_id": "span-6156d9c91a14695579ccbe8a", + "text": "We offer course bundles and custom training solutions tailored for companies." + }, + { + "span_id": "span-c5a288853decb5cadb64ea87", + "text": "Reach out to Louis at louis@towardsai.net ." + }, + { + "span_id": "span-64991fe8e7716bb8a0b86ccf", + "text": "If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days." + }, + { + "span_id": "span-a1c4e2ceb4e42c3da6804bef", + "text": "No hassle, no worries." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "a6bc8ca8d57344017821a9e2be227633bb1490aa061625319211457cf322834e", + "content_hash": "a6bc8ca8d57344017821a9e2be227633bb1490aa061625319211457cf322834e", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/bundles/from-coding-novice-to-advanced-llm-developer/" + ], + "evidence_hash": "9046f45efbaad348839f222ab6d19227fbc6f46ebc2f16fb1cbb2d94f4a8dafb" + }, + { + "discovered_url": "https://towardsai.com/academy/bundles/10-hour-crash-course-into-llm-developer-expert/", + "url": "https://towardsai.com/academy/bundles/10-hour-crash-course-into-llm-developer-expert/", + "canonical_url": "https://towardsai.com/academy/bundles/10-hour-crash-course-into-llm-developer-expert/", + "host": "towardsai.com", + "path": "/academy/bundles/10-hour-crash-course-into-llm-developer-expert", + "kind": "bundle", + "offer_id": "10-hour-crash-course-into-llm-developer-expert", + "entity_id": "offer:10-hour-crash-course-into-llm-developer-expert", + "title": "From Developer to Advanced AI Engineer · Course Bundle | Towards AI Academy", + "review_title": "From Developer to Advanced AI Engineer", + "reviewed_summary": "Bundle path from LLM fundamentals to advanced engineering.", + "meta_description": "Three courses and the e-book for $948: 10-Hour LLM Fundamentals, Full Stack AI Engineering, and Agent Engineering. Your roadmap to senior AI engineering capability.", + "headings": [ + "From developer to advanced AI engineer", + "How these courses work together", + "10-Hour LLM Fundamentals", + "Building LLMs for Production", + "Full Stack AI Engineering + Agent Engineering", + "Four products, one roadmap", + "Full Stack AI Engineering", + "Agent Engineering", + "Perfect for any developer to become an expert LLM developer", + "Frequently asked questions" + ], + "text": "Course bundle · for developers\n\nFrom developer to advanced AI engineer\nYour roadmap to becoming a senior AI engineer: three transformative courses and our e-book in one bundle. Start with the essentials and graduate to building advanced AI systems. $948 bundle price, was $1,078 4 products, fundamentals to agents 5.0 average course rating $1,078 $948 one-time · lifetime access 10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Agent Engineering ($499) Building LLMs for Production, the e-book ($29.99) Show me the full path 30-day money-back guarantee, no questions asked 01 The progression\n\nHow these courses work together\nThis is where theory meets practice, and where learners become experts. 01\n\n10-Hour LLM Fundamentals\nStart working with leading AI models like OpenAI, Gemini, and Claude. Adopt customization, databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows. Five two-hour sessions from basic prompt to production rollout. 02\n\nBuilding LLMs for Production\nLearn timeless principles for implementing advanced LLM techniques. Study the code logic behind prompting, RAG, fine-tuning, agents, and tools, and grasp LLM deployment and optimization strategies. 03\n\nFull Stack AI Engineering + Agent Engineering\nBuild advanced products and pipelines: production-ready implementations of RAG, fine-tuning, agents, and tools, mastered by building advanced AI projects and deployed multi-agent systems. 02 Bundle includes\n\nFour products, one roadmap\n$199 · rated 5.0\n\n10-Hour LLM Fundamentals\nTransform your ideas into AI-powered solutions. Master LLMs from prompts to deployment with the essential skills for production-ready applications. $349 · rated 5.0 (43 reviews)\n\nFull Stack AI Engineering\nBuild a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment. $499 · rated 5.0 (26 reviews)\n\nAgent Engineering\nFor engineers tasked with integrating autonomous features. Leave fully prepared to design, build, evaluate, and deploy production-grade AI agents. $29.99 · the e-book\n\nBuilding LLMs for Production\nTechniques that scale to enterprise workflows, helping developers and small teams with limited resources create sellable AI products. $948, was $1,078 · 30-day guarantee\n\nPerfect for any developer to become an expert LLM developer\nDive in with the 10-hour primer, learn timeless LLM principles with the e-book, and master production with the 60-hour Full Stack course plus Agent Engineering. Insights, practice, guidance, and a valuable portfolio project. Start risk-free Compare the courses 03 Questions\n\nFrequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries.", + "links": [ + { + "text": "Show me the full path", + "url": "https://academy.towardsai.net/cart/add_product/3251186?price_id=4144792" + }, + { + "text": "", + "url": "https://towardsai.com/academy/llm-primer/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/full-stack-ai-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/agent-engineering/" + }, + { + "text": "", + "url": "https://towardsai.com/academy/book/" + }, + { + "text": "Start risk-free", + "url": "https://academy.towardsai.net/cart/add_product/3251186?price_id=4144792" + }, + { + "text": "Compare the courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Affiliate page", + "url": "https://towardsai.com/academy/affiliate/" + } + ], + "chunks": [ + { + "chunk_id": "tai-09fa5172ec18e8aa1d847e0e", + "text": "Course bundle · for developers\n\nFrom developer to advanced AI engineer\nYour roadmap to becoming a senior AI engineer: three transformative courses and our e-book in one bundle. Start with the essentials and graduate to building advanced AI systems. $948 bundle price, was $1,078 4 products, fundamentals to agents 5.0 average course rating $1,078 $948 one-time · lifetime access 10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Agent Engineering ($499) Building LLMs for Production, the e-book ($29.99) Show me the full path 30-day money-back guarantee, no questions asked 01 The progression\n\nHow these courses work together\nThis is where theory meets practice, and where learners become experts. 01\n\n10-Hour LLM Fundamentals\nStart working with leading AI models like OpenAI, Gemini, and Claude. Adopt customization, databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows. Five two-hour sessions from basic prompt to production rollout. 02\n\nBuilding LLMs for Production\nLearn timeless principles for implementing advanced LLM techniques. Study the code logic behind prompting, RAG, fine-tuning, agents, and tools, and grasp LLM deployment and optimization strategies. 03\n\nFull Stack AI Engineering + Agent Engineering\nBuild advanced products and pipelines: production-ready implementations of RAG, fine-tuning, agents, and tools, mastered by building advanced AI projects and deployed multi-agent systems. 02 Bundle includes\n\nFour products, one roadmap\n$199 · rated 5.0\n\n10-Hour LLM Fundamentals\nTransform your ideas into AI-powered solutions. Master LLMs from prompts to deployment with the essential skills for production-ready applications. $349 · rated 5.0 (43 reviews)", + "heading": "From developer to advanced AI engineer", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-45876a1e2630c4c5565d305f", + "text": "Course bundle · for developers" + }, + { + "span_id": "span-fa1f4d8ed4590a3ded5847d1", + "text": "Your roadmap to becoming a senior AI engineer: three transformative courses and our e-book in one bundle." + }, + { + "span_id": "span-91a01f9dde499a1131cbe525", + "text": "Start with the essentials and graduate to building advanced AI systems." + }, + { + "span_id": "span-5829087eb0c29df7ae37d4d4", + "text": "$948 bundle price, was $1,078" + }, + { + "span_id": "span-7f833bceecbcfc27acd58623", + "text": "4 products, fundamentals to agents" + }, + { + "span_id": "span-54533e26fd1498feb15f9b77", + "text": "5.0 average course rating" + }, + { + "span_id": "span-dbff0112586ec068a0073cbc", + "text": "$1,078 $948 one-time · lifetime access" + }, + { + "span_id": "span-df0dd8f63fd6c95529ccee5b", + "text": "10-Hour LLM Fundamentals ($199) Full Stack AI Engineering ($349) Agent Engineering ($499) Building LLMs for Production, the e-book ($29.99)" + }, + { + "span_id": "span-6102c1a96e624e6a751210d1", + "text": "Show me the full path 30-day money-back guarantee, no questions asked" + }, + { + "span_id": "span-ec4a55c4a23ddd4544007c6f", + "text": "01 The progression" + }, + { + "span_id": "span-aaae8e06bd9373a4a5afd646", + "text": "This is where theory meets practice, and where learners become experts." + }, + { + "span_id": "span-e969b0f6e744e83e42bf4339", + "text": "10-Hour LLM Fundamentals" + }, + { + "span_id": "span-893851684f85292e85484f7e", + "text": "Start working with leading AI models like OpenAI, Gemini, and Claude." + }, + { + "span_id": "span-6632a7ee3ba8b13a4a5c88b6", + "text": "Adopt customization, databases, retrieval, frameworks, fine-tuning, and evaluations into your daily workflows." + }, + { + "span_id": "span-19a8c1a9cdec825de99f9deb", + "text": "Five two-hour sessions from basic prompt to production rollout." + }, + { + "span_id": "span-e7675223b8d772e8899d0f0e", + "text": "Learn timeless principles for implementing advanced LLM techniques." + }, + { + "span_id": "span-26362b4e98d878e71d28ad46", + "text": "Study the code logic behind prompting, RAG, fine-tuning, agents, and tools, and grasp LLM deployment and optimization strategies." + }, + { + "span_id": "span-47163d15887a77a9116ef4ab", + "text": "Build advanced products and pipelines: production-ready implementations of RAG, fine-tuning, agents, and tools, mastered by building advanced AI projects and deployed multi-agent systems." + }, + { + "span_id": "span-a8c44d321f001e91ea594f1b", + "text": "02 Bundle includes" + }, + { + "span_id": "span-ee6481b0d7889b3b31ec72c0", + "text": "$199 · rated 5.0" + }, + { + "span_id": "span-a4d244a5a14a7459e516a238", + "text": "Transform your ideas into AI-powered solutions." + }, + { + "span_id": "span-71f6c32aa534e77d252b5197", + "text": "Master LLMs from prompts to deployment with the essential skills for production-ready applications." + }, + { + "span_id": "span-db8b030430d556d0dba76a99", + "text": "$349 · rated 5.0 (43 reviews)" + } + ] + }, + { + "chunk_id": "tai-ab88c2490b6212988975b5ff", + "text": "Full Stack AI Engineering\nBuild a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment. $499 · rated 5.0 (26 reviews)\n\nAgent Engineering\nFor engineers tasked with integrating autonomous features. Leave fully prepared to design, build, evaluate, and deploy production-grade AI agents. $29.99 · the e-book\n\nBuilding LLMs for Production\nTechniques that scale to enterprise workflows, helping developers and small teams with limited resources create sellable AI products. $948, was $1,078 · 30-day guarantee\n\nPerfect for any developer to become an expert LLM developer\nDive in with the 10-hour primer, learn timeless LLM principles with the e-book, and master production with the 60-hour Full Stack course plus Agent Engineering. Insights, practice, guidance, and a valuable portfolio project. Start risk-free Compare the courses 03 Questions", + "heading": "Perfect for any developer to become an expert LLM developer", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-f5f6d320691c349fe712f034", + "text": "Build a working full-stack LLM product and master the skills companies need: context engineering, RAG, fine-tuning, agents, and deployment." + }, + { + "span_id": "span-3129e87786e814f64eca9c6c", + "text": "$499 · rated 5.0 (26 reviews)" + }, + { + "span_id": "span-1e4d421e1102045c8ba2712e", + "text": "For engineers tasked with integrating autonomous features." + }, + { + "span_id": "span-90fd1ec2cf9b2eefbda02165", + "text": "Leave fully prepared to design, build, evaluate, and deploy production-grade AI agents." + }, + { + "span_id": "span-ab22239f7160abf2af0e2586", + "text": "$29.99 · the e-book" + }, + { + "span_id": "span-900cad0350963c5aaa5c400b", + "text": "Techniques that scale to enterprise workflows, helping developers and small teams with limited resources create sellable AI products." + }, + { + "span_id": "span-1ec8ecf7431a0fe6bbd890e8", + "text": "$948, was $1,078 · 30-day guarantee" + }, + { + "span_id": "span-1ca0dfef9c339918c5ebde8f", + "text": "Dive in with the 10-hour primer, learn timeless LLM principles with the e-book, and master production with the 60-hour Full Stack course plus Agent Engineering." + }, + { + "span_id": "span-82f0eb6e56188061a5da54d0", + "text": "Insights, practice, guidance, and a valuable portfolio project." + }, + { + "span_id": "span-7716e09c5b8a1f1997452d89", + "text": "Start risk-free Compare the courses" + }, + { + "span_id": "span-bb62dcf68b331cd6d532f944", + "text": "03 Questions" + } + ] + }, + { + "chunk_id": "tai-01d3a28715522c5c766e2776", + "text": "Frequently asked questions\nWill I get a certificate upon completion? + Yes, you will receive a Towards AI certificate upon completion. When were these courses created? Are they still up to date? + The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves. You're getting a living resource designed to keep you ahead, not a static course. Can I get a discount? + Yes. Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more. Email Louis-François at louis@towardsai.net for a follow-up. Do you have a referral or affiliate program? + For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book. We also run an affiliate program for people with an audience. See the Affiliate page or email louis@towardsai.net with proof of referral. Can I take these courses within my company? + Yes. We offer course bundles and custom training solutions tailored for companies. Reach out to Louis at louis@towardsai.net . What is your refund policy? + If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days. No hassle, no worries.", + "heading": "Frequently asked questions", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-6fffaddaf9a7f41e12260ad9", + "text": "Yes, you will receive a Towards AI certificate upon completion." + }, + { + "span_id": "span-4890fb71e6b25e4d959a0fb6", + "text": "The courses originally launched in 2025, and we update them regularly with new lessons, tools, and techniques as the AI landscape evolves." + }, + { + "span_id": "span-ce93b7b7c9531b0cee32c594", + "text": "You're getting a living resource designed to keep you ahead, not a static course." + }, + { + "span_id": "span-b525e63872cbc336401f7405", + "text": "Students get 50% off all courses, previous book or course buyers are eligible for a discount, and we provide a bundle discount for groups of two or more." + }, + { + "span_id": "span-2c0d3461a3a5fd5e3c86b7d2", + "text": "Email Louis-François at louis@towardsai.net for a follow-up." + }, + { + "span_id": "span-ea28833a21ab62bc1407d79c", + "text": "For every referral you provide, you'll gain access to our exclusive eBook, and three or more referrals earn a physical copy of our book." + }, + { + "span_id": "span-5e10c160e6f91cc4395cdc57", + "text": "We also run an affiliate program for people with an audience." + }, + { + "span_id": "span-4cace2fed0543673c0ec8f12", + "text": "See the Affiliate page or email louis@towardsai.net with proof of referral." + }, + { + "span_id": "span-8f961da626bd8a56e1ba1b4c", + "text": "We offer course bundles and custom training solutions tailored for companies." + }, + { + "span_id": "span-8e8b3fc7446e0eff6f26ad0f", + "text": "Reach out to Louis at louis@towardsai.net ." + }, + { + "span_id": "span-14a48fe8b450690448d06334", + "text": "If, after completing the first part of a course, you feel it isn't right for you, we'll provide a 100% refund within 30 days." + }, + { + "span_id": "span-8f50d230854a04add8a3abfe", + "text": "No hassle, no worries." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "6d091614da07bf6e8e4893d6d43f39d053b980921a1aaafd56835aac48797ee7", + "content_hash": "6d091614da07bf6e8e4893d6d43f39d053b980921a1aaafd56835aac48797ee7", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/academy/bundles/10-hour-crash-course-into-llm-developer-expert/" + ], + "evidence_hash": "ab38b7294027b3ebee1aacc8da057ce8cad43a497ad22db6e6195961d0a77ae4" + }, + { + "discovered_url": "https://towardsai.com/enterpriseenablement/", + "url": "https://towardsai.com/enterpriseenablement/", + "canonical_url": "https://towardsai.com/enterpriseenablement/", + "host": "towardsai.com", + "path": "/enterpriseenablement", + "kind": "b2b", + "offer_id": null, + "entity_id": "page:towardsai.com/enterpriseenablement", + "title": "Enterprise AI Enablement · Towards AI", + "review_title": "Enterprise Enablement", + "reviewed_summary": "Enterprise AI training and enablement.", + "meta_description": "Towards AI helps enterprise teams move from AI pilots to production. Strategic advisory, training programmes for non-technical teams and engineers, prototype sprints, and full production builds.", + "headings": [ + "We teach you to build relevant AI tools.", + "AI fluency gaps are why most pilots never reach production.", + "Nine reasons AI pilots don’t reach production.", + "Wrong team lead", + "Weak domain expertise integration", + "“On-prem only” reflex", + "Optimising for cheap, not smart", + "Assuming staff “just get” AI", + "Workflow bolt-on", + "Demo-to-production gap", + "Evals and validation gap", + "Best methods stay private inside leading labs", + "Buy before build. Pods , not bodies.", + "Four ways to partner with us.", + "Prioritise use-cases by ROI, buy-vs-build, and system architecture.", + "Three delivery intensities .", + "Role-based learning paths for hundreds to thousands of seats.", + "An AI Excellence partner, on both sides of the build.", + "We guide", + "We build", + "We lead", + "How an engagement runs.", + "Scoping call, 60 minutes", + "Paid co-design", + "Pilot cohort", + "Certification and debrief", + "We scale to further waves only when the debrief earns it.", + "Best-in-class AI practitioners, distributed globally.", + "From non-technical to AI-powered operator .", + "Upskilling Europol’s special agents to use state-of-art AI tools to automate intelligence workflows.", + "Automating job-cost reporting for an HVAC services buy & build.", + "Tender discovery and prioritisation for an IT services buy & build.", + "Building an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when.", + "Tell us what your team needs to learn , improve , or build ." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Enterprise AI enablement\n\nWe teach you to build relevant AI tools.\nWe help enterprise teams move from AI pilots to custom production-ready deployments. Book a consultation Book a consultation See how we engage See how we engage Trusted by leading enterprises and institutions The gap we fix\n\nAI fluency gaps are why most pilots never reach production.\n95% of enterprise generative AI pilots fail to produce measurable P&L impact. MIT Sloan, 2025 5% of employees use AI in ways that meaningfully transform their work — despite 88% of organisations reporting AI adoption. EY Work Reimagined Survey, 15,000 employees, 2025 25% → 76% AI adoption jumps from 25% to 76% when employers provide structured AI training. Bright Horizons, 2025 60% of AI projects are unsupported by AI-ready data, driving low adoption. Gartner, 2025–26\n\nNine reasons AI pilots don’t reach production.\nAfter six years of building AI systems and training enterprise teams, we see the same pattern. 01\n\nWrong team lead\nPicking classic software or ML leaders instead of people with LLM and domain depth. 02\n\nWeak domain expertise integration\nEnd-users are often not properly embedded in workflow design. 03\n\n“On-prem only” reflex\nPrivacy and cost fears push teams onto small local language models that simply cannot do the job. 04\n\nOptimising for cheap, not smart\nAdoption happens at capability thresholds. A better model saves more human rework than it costs. 05\n\nAssuming staff “just get” AI\nMost people need ~20 hours of applied, hands-on training before they start to build safe, effective AI intuition. 06\n\nWorkflow bolt-on\nTools that do not reshape work produce no measurable P&L impact. High performers substantially redesign workflows. 07\n\nDemo-to-production gap\nSingle-turn happy-path demos hide that multi-step agents are distributed systems where every tool call can fail. 08\n\nEvals and validation gap\nBest-in-class AI teams spend more time on eval infrastructure than on model selection. 09\n\nBest methods stay private inside leading labs\nBy the time they hit blogs or open-source, you are a release cycle behind. You need teams generating novel approaches in-house. Sound familiar? Let’s talk about your bottlenecks. Book a consultation arrow_forward How we work\n\nBuy before build. Pods , not bodies.\nWe start with a buy-before-build ladder: an existing product plus training comes first, and custom software has to earn its place through reliability, integration, control, or unit cost. That discipline keeps budgets on the problems where bespoke engineering pays. When a build clears that bar, it runs through a small engineering pod: an AI engineer and a full-stack engineer paired with a deployment strategist, with architect support across pods. The pod that ships your system also trains the team that will own it. Every build follows one engineering rule. Code for rules. Models for meaning. People for responsibility.\n\nFour ways to partner with us.\nPick the engagement that matches where your team is. Hover any phase to see what's inside. 01 — Strategic Advisory arrow_forward 02 — AI Training Programmes arrow_forward 03 — Proof-of-concept Sprint arrow_forward 04 — Full Committed Development arrow_forward Advise\n\nPrioritise use-cases by ROI, buy-vs-build, and system architecture.\nModel selection and roadmaps that survive operating scrutiny. Helps leadership avoid the costly early mistakes that kill pilots. Use-case prioritisation tied to commercial impact Buy-vs-build, model selection, system architecture Roadmaps that hold up under operating scrutiny Training delivery\n\nThree delivery intensities .\nOne training system, delivered at the intensity your team needs. 01 — Enterprise Academy arrow_forward 02 — Team Bootcamps arrow_forward 03 — Towards AI Residency arrow_forward Per-seat licence\n\nRole-based learning paths for hundreds to thousands of seats.\nThe full course bundle with role-based paths, a live kickoff, monthly mentorship clinics, private cohort channels, and leadership analytics. Lifetime updates, because model guidance goes stale within months. Role-based paths across technical and non-technical teams Live kickoff, monthly clinics, private cohort channels Leadership analytics and lifetime content updates Beyond training\n\nAn AI Excellence partner, on both sides of the build.\n01\n\nWe guide\nAI strategy and standards: approved tools, review conventions, evaluation gates, and a prioritised use-case portfolio owned at leadership level. 02\n\nWe build\nCustom AI systems designed, shipped, and transferred by the same engineers who teach. 03\n\nWe lead\nFractional Chief AI Officer, offered as Fractional AI Leadership where the title sits better beside an existing CIO or CTO. Training is the entry point. We stay on as a long-term AI deployment partner. Working with us\n\nHow an engagement runs.\n01\n\nScoping call, 60 minutes\nWith the instructors who deliver the programme. There is no separate sales team. 02\n\nPaid co-design\nA nominee per priority team works with us to shape modules, exercises, and success measures. 03\n\nPilot cohort\nTypically 20 to 40 engineers. Real exercises from your roadmap, delivered on-site or remote, in English or French. 04\n\nCertification and debrief\nCertification against shipped capstones plus a written capability debrief per cohort. 05\n\nWe scale to further waves only when the debrief earns it.\nProcurement-ready: NDAs, PO and invoicing supported. No production access required. Your code stays yours. Technical team\n\nBest-in-class AI practitioners, distributed globally.\nDeployment strategists, forward-deployed engineers, and leaders from Palantir, BCG, UBS, J.P. Morgan, Aurelius Growth Capital, and MILA, working with AI engineers we trained in our own community. 01 🇨🇦 Canada AI Solutions Engineer LLM apps and integration 02 🇨🇦 Canada NLP Doctorate LLM researcher 03 🇮🇳 India Applied AI Engineer Agents specialist 04 🇨🇦 Canada Full-stack Engineer LLM integrator 05 🇩🇪 Germany Full-stack Engineer LLM developer 06 🇨🇦 Canada MILA-trained Engineer Retrieval-augmented generation 07 🇨🇦 Canada LLM Eval Engineer Fine-tuning specialist 08 🇲🇾 Malaysia AI Engineer Broad AI practitioner 09 🇺🇸 United States AI PhD Foundational models 10 🇨🇦 Canada AI Engineer MLOps, CV, NLP 11 🇮🇳 India AI Engineer Applied AI development 12 And many more Joining through our global community Case studies\n\nFrom non-technical to AI-powered operator .\nWe upskill, advise & build. Recent examples of our engagements. Enterprise training Paris\n\nUpskilling Europol’s special agents to use state-of-art AI tools to automate intelligence workflows.\nClient Europol — the law enforcement organisation that supports complex investigations and intelligence work across Europe. What we delivered Multi-day intensive bootcamp in Paris on advanced LLM pipelines, prompt engineering, and fine-tuning agents. Tailored to mission-critical law-enforcement workflows: multi-modal scraping, coded-language and slang identification. Secure deployment patterns for sensitive intelligence environments. Leading PE Fund Custom AI deployment USA\n\nAutomating job-cost reporting for an HVAC services buy & build.\nClient A leading PE fund’s HVAC services platform. Workers logged material costs per job on paper and backoffice re-typed them into Excel, growing days sales outstanding and weakening audit trails. What we delivered AI pipeline that reads PDFs of paper receipts. Automatic extraction and calculation of material costs per job. Real-time consolidation and job-level reporting. Audit-ready monthly reports, available instantly rather than at month-end. UK IT Services Group Custom AI deployment United Kingdom\n\nTender discovery and prioritisation for an IT services buy & build.\nClient A UK IT services group. Team members spent hours daily scanning fragmented tender sites across many geographies; high-value tenders were easy to miss and hard to prioritise. What we delivered Automated scraping across 5+ tender platforms. AI ranking by relevance to the group’s strengths and categorisation of likely win-rate. A daily report of newly-relevant opportunities, recovering many analyst hours per week. Nviya-Prime Agentic financial intelligence Europe\n\nBuilding an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when.\nClient Nviya-Prime, an agentic B2B financial-technology platform developed for a UK startup founded by a former banking partner. What we delivered An agentic intelligence platform combining filings, live news, risk data and future-event signals into prioritised, product-specific actions for bankers. Deterministic workflows and autonomous agents spanning three source tiers, with around 800 prompts and one million words of research per company. Modular, compliance-led deployment across Gemini, OpenAI, Claude and Grok, supporting client-cloud, API-only and locally hosted configurations. Book a consultation arrow_forward\n\nTell us what your team needs to learn , improve , or build .\nWe’ll help you prioritise the right use cases, train the right people, and ship the right systems. Every enquiry is read by the Towards AI leadership team. Book a consultation arrow_forward Email us Reply within one business day.", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation" + }, + { + "text": "Mentorship", + "url": "https://towardsai.com/academy/mentorship/" + }, + { + "text": "Discord", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Explore courses", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-c3844658cf8e1b54565e05bd", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Enterprise AI enablement\n\nWe teach you to build relevant AI tools.\nWe help enterprise teams move from AI pilots to custom production-ready deployments. Book a consultation Book a consultation See how we engage See how we engage Trusted by leading enterprises and institutions The gap we fix\n\nAI fluency gaps are why most pilots never reach production.\n95% of enterprise generative AI pilots fail to produce measurable P&L impact. MIT Sloan, 2025 5% of employees use AI in ways that meaningfully transform their work — despite 88% of organisations reporting AI adoption. EY Work Reimagined Survey, 15,000 employees, 2025 25% → 76% AI adoption jumps from 25% to 76% when employers provide structured AI training. Bright Horizons, 2025 60% of AI projects are unsupported by AI-ready data, driving low adoption. Gartner, 2025–26\n\nNine reasons AI pilots don’t reach production.\nAfter six years of building AI systems and training enterprise teams, we see the same pattern. 01\n\nWrong team lead\nPicking classic software or ML leaders instead of people with LLM and domain depth. 02\n\nWeak domain expertise integration\nEnd-users are often not properly embedded in workflow design. 03", + "heading": "AI fluency gaps are why most pilots never reach production.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-3fc1bf0f90bc4ba73908e563", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-4d85e31f8aa3f94f70ffdc43", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-c95b434ce05ea2c8fc450b83", + "text": "Community keyboard_arrow_down Mentorship Discord" + }, + { + "span_id": "span-78e9b98ac7589fd70c86db80", + "text": "Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-4ec5b3412484b784fe4b0589", + "text": "Explore courses Talk to us" + }, + { + "span_id": "span-073aa61d3ab5b58d16042518", + "text": "Enterprise AI enablement" + }, + { + "span_id": "span-0b650aebc1ef54d9d6257c2d", + "text": "We help enterprise teams move from AI pilots to custom production-ready deployments." + }, + { + "span_id": "span-febfd2aa8792712495621fa8", + "text": "Book a consultation Book a consultation See how we engage See how we engage" + }, + { + "span_id": "span-e95c2193bda32f690745576d", + "text": "Trusted by leading enterprises" + }, + { + "span_id": "span-0a039bd4478a7000bb43b8ec", + "text": "and institutions" + }, + { + "span_id": "span-1e4ecb3039a0b44d1a8b4886", + "text": "The gap we fix" + }, + { + "span_id": "span-617fb2cd61e363ae917a6b07", + "text": "of enterprise generative AI pilots fail to produce measurable P&L impact." + }, + { + "span_id": "span-04c9d406c237cec9a0f69098", + "text": "MIT Sloan, 2025" + }, + { + "span_id": "span-ceb9b74f2560af6d014450b4", + "text": "of employees use AI in ways that meaningfully transform their work — despite 88% of organisations reporting AI adoption." + }, + { + "span_id": "span-d07f94417db2e95429003dfc", + "text": "EY Work Reimagined Survey, 15,000 employees, 2025" + }, + { + "span_id": "span-6e51a920b496ea961219046a", + "text": "25% → 76%" + }, + { + "span_id": "span-eee1b80c29fc964d2985dd8b", + "text": "AI adoption jumps from 25% to 76% when employers provide structured AI training." + }, + { + "span_id": "span-f577a05a117cb6b2751e7cea", + "text": "Bright Horizons, 2025" + }, + { + "span_id": "span-ac5392d2b675a8e75d71d84e", + "text": "of AI projects are unsupported by AI-ready data, driving low adoption." + }, + { + "span_id": "span-c5d696df96479615f3657a08", + "text": "Gartner, 2025–26" + }, + { + "span_id": "span-212a8e9482b6bee929d07066", + "text": "After six years of building AI systems and training enterprise teams, we see the same pattern." + }, + { + "span_id": "span-42176248cacbd65c0b15bed1", + "text": "Picking classic software or ML leaders instead of people with LLM and domain depth." + }, + { + "span_id": "span-463d26a08ccb38d167dd8255", + "text": "End-users are often not properly embedded in workflow design." + } + ] + }, + { + "chunk_id": "tai-3c0fb402d76d3e5f61c254a5", + "text": "“On-prem only” reflex\nPrivacy and cost fears push teams onto small local language models that simply cannot do the job. 04\n\nOptimising for cheap, not smart\nAdoption happens at capability thresholds. A better model saves more human rework than it costs. 05\n\nAssuming staff “just get” AI\nMost people need ~20 hours of applied, hands-on training before they start to build safe, effective AI intuition. 06\n\nWorkflow bolt-on\nTools that do not reshape work produce no measurable P&L impact. High performers substantially redesign workflows. 07\n\nDemo-to-production gap\nSingle-turn happy-path demos hide that multi-step agents are distributed systems where every tool call can fail. 08\n\nEvals and validation gap\nBest-in-class AI teams spend more time on eval infrastructure than on model selection. 09\n\nBest methods stay private inside leading labs\nBy the time they hit blogs or open-source, you are a release cycle behind. You need teams generating novel approaches in-house. Sound familiar? Let’s talk about your bottlenecks. Book a consultation arrow_forward How we work\n\nBuy before build. Pods , not bodies.\nWe start with a buy-before-build ladder: an existing product plus training comes first, and custom software has to earn its place through reliability, integration, control, or unit cost. That discipline keeps budgets on the problems where bespoke engineering pays. When a build clears that bar, it runs through a small engineering pod: an AI engineer and a full-stack engineer paired with a deployment strategist, with architect support across pods. The pod that ships your system also trains the team that will own it. Every build follows one engineering rule. Code for rules. Models for meaning. People for responsibility.", + "heading": "Buy before build. Pods , not bodies.", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-6de63eae3f3ee74f84a095cf", + "text": "Privacy and cost fears push teams onto small local language models that simply cannot do the job." + }, + { + "span_id": "span-6079fc2adc950f1f72e51347", + "text": "Adoption happens at capability thresholds." + }, + { + "span_id": "span-11849db6f8d26a3b7a247528", + "text": "A better model saves more human rework than it costs." + }, + { + "span_id": "span-754d629ef0e85ddc211e710e", + "text": "Most people need ~20 hours of applied, hands-on training before they start to build safe, effective AI intuition." + }, + { + "span_id": "span-fd0fa446d2f61472a38284b2", + "text": "Tools that do not reshape work produce no measurable P&L impact." + }, + { + "span_id": "span-6589039a0fedefaf11782e02", + "text": "High performers substantially redesign workflows." + }, + { + "span_id": "span-bb461897ec8343a86c0b0c30", + "text": "Single-turn happy-path demos hide that multi-step agents are distributed systems where every tool call can fail." + }, + { + "span_id": "span-bd5a321b8236935f271ea668", + "text": "Best-in-class AI teams spend more time on eval infrastructure than on model selection." + }, + { + "span_id": "span-4fe8c49180f84e0e5907c93d", + "text": "By the time they hit blogs or open-source, you are a release cycle behind." + }, + { + "span_id": "span-574dd64a091f0965c05e2ce7", + "text": "You need teams generating novel approaches in-house." + }, + { + "span_id": "span-8bcbbc82828672b4405d7a50", + "text": "Sound familiar?" + }, + { + "span_id": "span-6ba57a4db9ff0bae38bc9774", + "text": "Let’s talk about your bottlenecks." + }, + { + "span_id": "span-6a2bfb1f92457c5fe147dcb5", + "text": "Book a consultation arrow_forward" + }, + { + "span_id": "span-762449e37c9af1a46ef28c55", + "text": "How we work" + }, + { + "span_id": "span-b36ea2d42e8fde19e1db48a9", + "text": "We start with a buy-before-build ladder: an existing product plus training comes first, and custom software has to earn its place through reliability, integration, control, or unit cost." + }, + { + "span_id": "span-7e50671d44b0c55bcfaf53ad", + "text": "That discipline keeps budgets on the problems where bespoke engineering pays." + }, + { + "span_id": "span-74e1edaf81c229c58f82cf86", + "text": "When a build clears that bar, it runs through a small engineering pod: an AI engineer and a full-stack engineer paired with a deployment strategist, with architect support across pods." + }, + { + "span_id": "span-d22119b672c3bdcef0a890b1", + "text": "The pod that ships your system also trains the team that will own it." + }, + { + "span_id": "span-99ef63f838bf80e2383deed4", + "text": "Every build follows one engineering rule." + }, + { + "span_id": "span-f71193ad3f870ca0f1b465ac", + "text": "Code for rules." + }, + { + "span_id": "span-7086f5d6d3e0967885785b95", + "text": "Models for meaning." + }, + { + "span_id": "span-3be4897f7bf8634ec9052a25", + "text": "People for responsibility." + } + ] + }, + { + "chunk_id": "tai-8376db1bfffc27e3378c4f1d", + "text": "Four ways to partner with us.\nPick the engagement that matches where your team is. Hover any phase to see what's inside. 01 — Strategic Advisory arrow_forward 02 — AI Training Programmes arrow_forward 03 — Proof-of-concept Sprint arrow_forward 04 — Full Committed Development arrow_forward Advise\n\nPrioritise use-cases by ROI, buy-vs-build, and system architecture.\nModel selection and roadmaps that survive operating scrutiny. Helps leadership avoid the costly early mistakes that kill pilots. Use-case prioritisation tied to commercial impact Buy-vs-build, model selection, system architecture Roadmaps that hold up under operating scrutiny Training delivery\n\nThree delivery intensities .\nOne training system, delivered at the intensity your team needs. 01 — Enterprise Academy arrow_forward 02 — Team Bootcamps arrow_forward 03 — Towards AI Residency arrow_forward Per-seat licence\n\nRole-based learning paths for hundreds to thousands of seats.\nThe full course bundle with role-based paths, a live kickoff, monthly mentorship clinics, private cohort channels, and leadership analytics. Lifetime updates, because model guidance goes stale within months. Role-based paths across technical and non-technical teams Live kickoff, monthly clinics, private cohort channels Leadership analytics and lifetime content updates Beyond training\n\nAn AI Excellence partner, on both sides of the build.\n01\n\nWe guide\nAI strategy and standards: approved tools, review conventions, evaluation gates, and a prioritised use-case portfolio owned at leadership level. 02", + "heading": "Role-based learning paths for hundreds to thousands of seats.", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-506078b263b8d08e4620353b", + "text": "Pick the engagement that matches where your team is." + }, + { + "span_id": "span-e08d903d0f5f18122ceec60d", + "text": "Hover any phase to see what's inside." + }, + { + "span_id": "span-993ff959c4d1bbd0b0bd24b8", + "text": "01 — Strategic Advisory arrow_forward 02 — AI Training Programmes arrow_forward 03 — Proof-of-concept Sprint arrow_forward 04 — Full Committed Development arrow_forward" + }, + { + "span_id": "span-fe2b006562b5e722245662d1", + "text": "Model selection and roadmaps that survive operating scrutiny." + }, + { + "span_id": "span-b364779c4e79e32ba6c3005f", + "text": "Helps leadership avoid the costly early mistakes that kill pilots." + }, + { + "span_id": "span-f2514ccec0aa08edabed48c2", + "text": "Use-case prioritisation tied to commercial impact" + }, + { + "span_id": "span-723a2759a0b3bbf87dc5ed88", + "text": "Buy-vs-build, model selection, system architecture" + }, + { + "span_id": "span-21e5063eebc9a151b15e65b5", + "text": "Roadmaps that hold up under operating scrutiny" + }, + { + "span_id": "span-04ac665d372ba729283390ba", + "text": "Training delivery" + }, + { + "span_id": "span-be49832570e7872f09f9b5cb", + "text": "One training system, delivered at the intensity your team needs." + }, + { + "span_id": "span-9b90ca79adf2bcbffc685dd7", + "text": "01 — Enterprise Academy arrow_forward 02 — Team Bootcamps arrow_forward 03 — Towards AI Residency arrow_forward" + }, + { + "span_id": "span-00191893a732aed6495dae0c", + "text": "Per-seat licence" + }, + { + "span_id": "span-895aeebcd51d1b11058be8d7", + "text": "The full course bundle with role-based paths, a live kickoff, monthly mentorship clinics, private cohort channels, and leadership analytics." + }, + { + "span_id": "span-c042668dd20a7dae9f959084", + "text": "Lifetime updates, because model guidance goes stale within months." + }, + { + "span_id": "span-077cad0694e587beec02474a", + "text": "Role-based paths across technical and non-technical teams" + }, + { + "span_id": "span-156aabc264a0ccdcaa08fbca", + "text": "Live kickoff, monthly clinics, private cohort channels" + }, + { + "span_id": "span-f856329c938aea400844564f", + "text": "Leadership analytics and lifetime content updates" + }, + { + "span_id": "span-654d84b2d08daec093202c77", + "text": "Beyond training" + }, + { + "span_id": "span-ffb0275d7b4444ed064198b0", + "text": "An AI Excellence partner, on both sides of the build." + }, + { + "span_id": "span-59456bc5d1f7edeaa3e2d83d", + "text": "AI strategy and standards: approved tools, review conventions, evaluation gates, and a prioritised use-case portfolio owned at leadership level." + } + ] + }, + { + "chunk_id": "tai-033d7c25d7b260b9fdf9ef37", + "text": "We build\nCustom AI systems designed, shipped, and transferred by the same engineers who teach. 03\n\nWe lead\nFractional Chief AI Officer, offered as Fractional AI Leadership where the title sits better beside an existing CIO or CTO. Training is the entry point. We stay on as a long-term AI deployment partner. Working with us\n\nHow an engagement runs.\n01\n\nScoping call, 60 minutes\nWith the instructors who deliver the programme. There is no separate sales team. 02\n\nPaid co-design\nA nominee per priority team works with us to shape modules, exercises, and success measures. 03\n\nPilot cohort\nTypically 20 to 40 engineers. Real exercises from your roadmap, delivered on-site or remote, in English or French. 04\n\nCertification and debrief\nCertification against shipped capstones plus a written capability debrief per cohort. 05\n\nWe scale to further waves only when the debrief earns it.\nProcurement-ready: NDAs, PO and invoicing supported. No production access required. Your code stays yours. Technical team", + "heading": "We lead", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-96c6e28d9d3e074e594c77df", + "text": "Custom AI systems designed, shipped, and transferred by the same engineers who teach." + }, + { + "span_id": "span-90d9b613beaf3afdfcf040f6", + "text": "Fractional Chief AI Officer, offered as Fractional AI Leadership where the title sits better beside an existing CIO or CTO." + }, + { + "span_id": "span-87708a7e23afd0dad4c9d4a5", + "text": "Training is the entry point." + }, + { + "span_id": "span-1ec65fbeb76c1d0fa7783cfc", + "text": "We stay on as a long-term AI deployment partner." + }, + { + "span_id": "span-162a573d5be2ccf8f525b418", + "text": "Working with us" + }, + { + "span_id": "span-e3d2c0f7168e39eff1a4bcfd", + "text": "How an engagement runs." + }, + { + "span_id": "span-ec0e4309b450a232f58d5d89", + "text": "Scoping call, 60 minutes" + }, + { + "span_id": "span-da4d5316bdf942801b9ac5f0", + "text": "With the instructors who deliver the programme." + }, + { + "span_id": "span-f5a068c50d65011d1d93b337", + "text": "There is no separate sales team." + }, + { + "span_id": "span-16db9430ff37a64224400cfa", + "text": "A nominee per priority team works with us to shape modules, exercises, and success measures." + }, + { + "span_id": "span-fed1d2bc3a30740113377600", + "text": "Typically 20 to 40 engineers." + }, + { + "span_id": "span-743d4d34d452c2990fe00e6e", + "text": "Real exercises from your roadmap, delivered on-site or remote, in English or French." + }, + { + "span_id": "span-bc44c636b2aed302ae18472d", + "text": "Certification against shipped capstones plus a written capability debrief per cohort." + }, + { + "span_id": "span-8294d6f18435334cce012f93", + "text": "Procurement-ready: NDAs, PO and invoicing supported." + }, + { + "span_id": "span-9ebd776d95baced6c0e3784d", + "text": "No production access required." + }, + { + "span_id": "span-0a59a1271ef0e758acffbb4e", + "text": "Your code stays yours." + }, + { + "span_id": "span-edd66bb2ce27bd062cb97abf", + "text": "Technical team" + } + ] + }, + { + "chunk_id": "tai-b59dfba7c192bd03fee1d4c9", + "text": "Best-in-class AI practitioners, distributed globally.\nDeployment strategists, forward-deployed engineers, and leaders from Palantir, BCG, UBS, J.P. Morgan, Aurelius Growth Capital, and MILA, working with AI engineers we trained in our own community. 01 🇨🇦 Canada AI Solutions Engineer LLM apps and integration 02 🇨🇦 Canada NLP Doctorate LLM researcher 03 🇮🇳 India Applied AI Engineer Agents specialist 04 🇨🇦 Canada Full-stack Engineer LLM integrator 05 🇩🇪 Germany Full-stack Engineer LLM developer 06 🇨🇦 Canada MILA-trained Engineer Retrieval-augmented generation 07 🇨🇦 Canada LLM Eval Engineer Fine-tuning specialist 08 🇲🇾 Malaysia AI Engineer Broad AI practitioner 09 🇺🇸 United States AI PhD Foundational models 10 🇨🇦 Canada AI Engineer MLOps, CV, NLP 11 🇮🇳 India AI Engineer Applied AI development 12 And many more Joining through our global community Case studies\n\nFrom non-technical to AI-powered operator .\nWe upskill, advise & build. Recent examples of our engagements. Enterprise training Paris\n\nUpskilling Europol’s special agents to use state-of-art AI tools to automate intelligence workflows.\nClient Europol — the law enforcement organisation that supports complex investigations and intelligence work across Europe. What we delivered Multi-day intensive bootcamp in Paris on advanced LLM pipelines, prompt engineering, and fine-tuning agents. Tailored to mission-critical law-enforcement workflows: multi-modal scraping, coded-language and slang identification. Secure deployment patterns for sensitive intelligence environments. Leading PE Fund Custom AI deployment USA", + "heading": "Best-in-class AI practitioners, distributed globally.", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-13b03c90c0b025d20e1aa301", + "text": "Deployment strategists, forward-deployed engineers, and leaders from Palantir, BCG, UBS, J.P." + }, + { + "span_id": "span-8faf90b49de94ba98bec3b26", + "text": "Morgan, Aurelius Growth Capital, and MILA, working with AI engineers we trained in our own community." + }, + { + "span_id": "span-e82230279dbf9ab72e5fefe5", + "text": "AI Solutions Engineer" + }, + { + "span_id": "span-f2414cad726147d9dbb08b79", + "text": "LLM apps and integration" + }, + { + "span_id": "span-7e091927ab75183aff7ae230", + "text": "NLP Doctorate" + }, + { + "span_id": "span-07d30d45d67ce6815843ed1c", + "text": "LLM researcher" + }, + { + "span_id": "span-426fd929aa7cd8ea849544fa", + "text": "Applied AI Engineer" + }, + { + "span_id": "span-e7d697a1db29ec774fb12441", + "text": "Agents specialist" + }, + { + "span_id": "span-7335f46e3ddc4ddc8b8dded8", + "text": "Full-stack Engineer" + }, + { + "span_id": "span-7869dacd7c0524f490f4cafe", + "text": "LLM integrator" + }, + { + "span_id": "span-c736bb995703b8997159d1a5", + "text": "LLM developer" + }, + { + "span_id": "span-71085f06babba73b39fc9ae7", + "text": "MILA-trained Engineer" + }, + { + "span_id": "span-60f233ba9fbcede3f1843993", + "text": "Retrieval-augmented generation" + }, + { + "span_id": "span-9714bdd11ee526bb0637755e", + "text": "LLM Eval Engineer" + }, + { + "span_id": "span-0027324720b3d2f8133e4fe2", + "text": "Fine-tuning specialist" + }, + { + "span_id": "span-c86844f5ca20b1f442b9241b", + "text": "AI Engineer" + }, + { + "span_id": "span-e0190aafaa00e6d159baac4a", + "text": "Broad AI practitioner" + }, + { + "span_id": "span-eda11fa84e68b81d7df8b7c3", + "text": "United States" + }, + { + "span_id": "span-31498fb7e36013b65b59060a", + "text": "AI PhD" + }, + { + "span_id": "span-8bb4b35ab304d60ca6d1de01", + "text": "Foundational models" + }, + { + "span_id": "span-628f796df831fb06ea8281e0", + "text": "MLOps, CV, NLP" + }, + { + "span_id": "span-16faa318102bd77a06ac5a8b", + "text": "Applied AI development" + }, + { + "span_id": "span-8b81104b87492821974bac43", + "text": "And many more" + }, + { + "span_id": "span-0cca5d096ce3960e53dc092b", + "text": "Joining through our global community" + }, + { + "span_id": "span-c2d344f45e84b986b92213ae", + "text": "Case studies" + }, + { + "span_id": "span-9ca135688799da5fd2faa233", + "text": "We upskill, advise & build." + }, + { + "span_id": "span-f6de765de8f4388fa94c972e", + "text": "Recent examples of our engagements." + }, + { + "span_id": "span-cd5f1b694ad6e8cbd8a0e1fd", + "text": "Enterprise training Paris" + }, + { + "span_id": "span-54ef9ed7edcb79a1627a8f1b", + "text": "Europol — the law enforcement organisation that supports complex investigations and intelligence work across Europe." + }, + { + "span_id": "span-35b53022ca88127bbba33743", + "text": "What we delivered" + }, + { + "span_id": "span-176751f8a68cacb9c8784725", + "text": "Multi-day intensive bootcamp in Paris on advanced LLM pipelines, prompt engineering, and fine-tuning agents." + }, + { + "span_id": "span-0615da52a395586b5c06d52c", + "text": "Tailored to mission-critical law-enforcement workflows: multi-modal scraping, coded-language and slang identification." + }, + { + "span_id": "span-c66e67ff6831a330f6636cea", + "text": "Secure deployment patterns for sensitive intelligence environments." + }, + { + "span_id": "span-10b07d50efc3170dc1e67044", + "text": "Leading PE Fund Custom AI deployment USA" + } + ] + }, + { + "chunk_id": "tai-ae8956290096c1ff911f2506", + "text": "Automating job-cost reporting for an HVAC services buy & build.\nClient A leading PE fund’s HVAC services platform. Workers logged material costs per job on paper and backoffice re-typed them into Excel, growing days sales outstanding and weakening audit trails. What we delivered AI pipeline that reads PDFs of paper receipts. Automatic extraction and calculation of material costs per job. Real-time consolidation and job-level reporting. Audit-ready monthly reports, available instantly rather than at month-end. UK IT Services Group Custom AI deployment United Kingdom\n\nTender discovery and prioritisation for an IT services buy & build.\nClient A UK IT services group. Team members spent hours daily scanning fragmented tender sites across many geographies; high-value tenders were easy to miss and hard to prioritise. What we delivered Automated scraping across 5+ tender platforms. AI ranking by relevance to the group’s strengths and categorisation of likely win-rate. A daily report of newly-relevant opportunities, recovering many analyst hours per week. Nviya-Prime Agentic financial intelligence Europe", + "heading": "Automating job-cost reporting for an HVAC services buy & build.", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-8bb2b99ba4c6804876df14b4", + "text": "A leading PE fund’s HVAC services platform." + }, + { + "span_id": "span-827692b839ff56432b0a220f", + "text": "Workers logged material costs per job on paper and backoffice re-typed them into Excel, growing days sales outstanding and weakening audit trails." + }, + { + "span_id": "span-51acffdc38183c57e12124fe", + "text": "What we delivered" + }, + { + "span_id": "span-e5d50527ad90c29adab2ca97", + "text": "AI pipeline that reads PDFs of paper receipts." + }, + { + "span_id": "span-6f74007ef6bc542279efb280", + "text": "Automatic extraction and calculation of material costs per job." + }, + { + "span_id": "span-b69f7d008620e014f1d417b2", + "text": "Real-time consolidation and job-level reporting." + }, + { + "span_id": "span-edc707dd8afdeb79668192a2", + "text": "Audit-ready monthly reports, available instantly rather than at month-end." + }, + { + "span_id": "span-765436345b7264f85b96811f", + "text": "UK IT Services Group Custom AI deployment United Kingdom" + }, + { + "span_id": "span-477d54c1b0e2075bcc9e88ad", + "text": "A UK IT services group." + }, + { + "span_id": "span-6d2d331559a5fc62bdb835e6", + "text": "Team members spent hours daily scanning fragmented tender sites across many geographies; high-value tenders were easy to miss and hard to prioritise." + }, + { + "span_id": "span-6d88a0cb0e4ba0bcb92ecc31", + "text": "Automated scraping across 5+ tender platforms." + }, + { + "span_id": "span-dcd20c02f648c4b32ac448b9", + "text": "AI ranking by relevance to the group’s strengths and categorisation of likely win-rate." + }, + { + "span_id": "span-c672ab28ea3eeb2dd3f326fd", + "text": "A daily report of newly-relevant opportunities, recovering many analyst hours per week." + }, + { + "span_id": "span-e0a3e0f8e8c2355ea6df3563", + "text": "Nviya-Prime Agentic financial intelligence Europe" + } + ] + }, + { + "chunk_id": "tai-df038b4c1ef4af788dc4d609", + "text": "Building an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when.\nClient Nviya-Prime, an agentic B2B financial-technology platform developed for a UK startup founded by a former banking partner. What we delivered An agentic intelligence platform combining filings, live news, risk data and future-event signals into prioritised, product-specific actions for bankers. Deterministic workflows and autonomous agents spanning three source tiers, with around 800 prompts and one million words of research per company. Modular, compliance-led deployment across Gemini, OpenAI, Claude and Grok, supporting client-cloud, API-only and locally hosted configurations. Book a consultation arrow_forward\n\nTell us what your team needs to learn , improve , or build .\nWe’ll help you prioritise the right use cases, train the right people, and ship the right systems. Every enquiry is read by the Towards AI leadership team. Book a consultation arrow_forward Email us Reply within one business day.", + "heading": "Building an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when.", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-5202f585019d04d9cdc16033", + "text": "Nviya-Prime, an agentic B2B financial-technology platform developed for a UK startup founded by a former banking partner." + }, + { + "span_id": "span-2b62b85b8b707dcef3ce20a0", + "text": "What we delivered" + }, + { + "span_id": "span-d8dc1ab59deed3b715530fbc", + "text": "An agentic intelligence platform combining filings, live news, risk data and future-event signals into prioritised, product-specific actions for bankers." + }, + { + "span_id": "span-2c6b7ce649c1166842c8167a", + "text": "Deterministic workflows and autonomous agents spanning three source tiers, with around 800 prompts and one million words of research per company." + }, + { + "span_id": "span-91957a6361b9ab4e3924822c", + "text": "Modular, compliance-led deployment across Gemini, OpenAI, Claude and Grok, supporting client-cloud, API-only and locally hosted configurations." + }, + { + "span_id": "span-6d99355db446ca2904fc50df", + "text": "Book a consultation arrow_forward" + }, + { + "span_id": "span-e3658548ce2cbd46643e23e5", + "text": "We’ll help you prioritise the right use cases, train the right people, and ship the right systems." + }, + { + "span_id": "span-c39869f9929f33831afd3391", + "text": "Every enquiry is read by the Towards AI leadership team." + }, + { + "span_id": "span-fe00011a032c9a8853af090d", + "text": "Book a consultation arrow_forward Email us" + }, + { + "span_id": "span-614a536f02bcb7c53e6e3986", + "text": "Reply within one business day." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "5265141ebe7d3804fdf9384e9927e7ed28d1478bec6dc881afafc9ef5b855b5c", + "content_hash": "5265141ebe7d3804fdf9384e9927e7ed28d1478bec6dc881afafc9ef5b855b5c", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/enterpriseenablement/" + ], + "evidence_hash": "fe357cb2a77e1b3eb2804bda4199b8780079e6193b255e78cb55101a6573790f" + }, + { + "discovered_url": "https://towardsai.com/valuecreation/", + "url": "https://towardsai.com/valuecreation/", + "canonical_url": "https://towardsai.com/valuecreation/", + "host": "towardsai.com", + "path": "/valuecreation", + "kind": "b2b", + "offer_id": null, + "entity_id": "page:towardsai.com/valuecreation", + "title": "Custom AI Development & PE Value Creation · Towards AI", + "review_title": "Value Creation", + "reviewed_summary": "Custom AI development and value-creation consulting.", + "meta_description": "Towards AI deploys production-ready AI directly into PE portfolio companies, tied to EBITDA levers, not strategy decks. Custom AI systems for revenue acceleration, operational automation, and cost-to-serve reduction.", + "headings": [ + "Engineering Alpha .", + "From AI enablement to sustained value creation .", + "AI-fluent teams, before any build.", + "Three things. Rarely found together.", + "Finance expertise.", + "AI expertise.", + "Deployment expertise.", + "Recent engagements .", + "Every portfolio company has an EBITDA lever . Let’s find yours." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us\n\nEngineering Alpha .\nProprietary AI playbooks , stress-tested in PE-backed environments to move real EBITDA . Request a briefing Request a briefing See our work See our work Built by PE investors, operators, AI engineers\n\nFrom AI enablement to sustained value creation .\nWe offer enablement + 3 engagement types, sold standalone or in phases, leaving every team more AI-native than it found them. 00 — Enablement arrow_forward 01 — Scope arrow_forward 02 — Build arrow_forward 03 — Sustain arrow_forward Foundations · Runs across all phases\n\nAI-fluent teams, before any build.\nSet the foundations so non-technical staff, engineers, and leadership all speak the same AI language. Most of our work lands harder when these are already in place. Foundations: AI policy, governance, and responsible-use guardrails Non-technical teams: 10x productivity, AI-native ways of working Engineering teams: best-in-class full-stack AI engineers Leadership: PE partners, operating partners, C-level\n\nThree things. Rarely found together.\nMost AI firms are technical but don’t understand value creation. Most consultants understand value creation but lack AI expertise. We bring both, with proven playbooks for AI deployments tied to value creation.\n\nFinance expertise.\nOur co-founders come from J.P. Morgan, where they led trading, credit research, and asset management, and from Aurelius, a leading buy-and-build PE fund. We model EBITDA levers the way a finance team would, not the way an AI lab would.\n\nAI expertise.\nOur co-founders are recognised AI thought leaders with 200k+ followers and a best-selling book, Building LLMs for Production, used as a reference inside enterprise AI teams. 10,000+ copies sold.\n\nDeployment expertise.\nOur Towards AI Deployment practice ships custom systems through engineering pods, with strategists and leaders from Palantir, BCG, UBS, J.P. Morgan, Aurelius Growth Capital, and MILA. We know how AI lands in real organisations because we have been the people landing it.\n\nRecent engagements .\nSaved hours are capacity, not EBITDA. We count impact only when the operating plan changes. 🇺🇸 Leading PE Fund HVAC services buy & build · USA Problem statement Workers logged material costs per job on paper. Backoffice manually re-typed from paper into Excel. Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails. What we built AI pipeline that takes PDFs of paper receipts Automatic extraction & calculation of material costs per Job Real-time consolidation & job-level reporting Audit-ready monthly reports Outcome Hours of finance team time saved per week. Audit-ready reports available instantly, not at month-end. 🇬🇧 UK IT Services Group IT services buy & build · UK Problem statement Multiple team members spent hours daily scanning fragmented tender sites across many geographies. Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most. What we built Automated scraping across 5+ tender platforms AI ranking by relevance to the group’s strengths AI categorization of likely win-rate Daily report of newly-relevant opportunities Outcome Many hours of analyst time recovered per week. Pipeline focused on the tenders the group is structurally most likely to win. 🇫🇷 Europol Law enforcement agency · France Problem statement Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines. What we built Multi-day intensive bootcamp on LLM pipelines Tailored to law-enforcement workflows Secure deployment for sensitive intelligence work Hands-on labs for context engineering Outcome Special agents trained and ready to deploy AI tools across operational intelligence workflows. 🇪🇺 Nviya-Prime Agentic financial intelligence · Europe The brief Build an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when, for an agentic B2B financial-technology platform founded by a former banking partner. What we delivered Agentic platform combining filings, live news, risk data and future-event signals Prioritised, product-specific pitch actions for bankers Around 800 prompts and one million words of research per company Compliance-led deployment across Gemini, OpenAI, Claude and Grok Outcome Banker reasoning reproduced at scale across client-cloud, API-only and locally hosted configurations, in live demonstrations with global banks.\n\nEvery portfolio company has an EBITDA lever . Let’s find yours.\nA 30-minute call with the Towards AI leadership team. We’ll discuss the value-creation priorities on your desk, where AI can move the number, and the shape of an engagement. Request a briefing arrow_forward Email us Reply within one business day.", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation" + }, + { + "text": "Mentorship", + "url": "https://towardsai.com/academy/mentorship/" + }, + { + "text": "Discord", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Explore courses", + "url": "https://towardsai.com/academy/" + } + ], + "chunks": [ + { + "chunk_id": "tai-63c539b4b3f49ef11ad7478e", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community keyboard_arrow_down Mentorship Discord Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us\n\nEngineering Alpha .\nProprietary AI playbooks , stress-tested in PE-backed environments to move real EBITDA . Request a briefing Request a briefing See our work See our work Built by PE investors, operators, AI engineers\n\nFrom AI enablement to sustained value creation .\nWe offer enablement + 3 engagement types, sold standalone or in phases, leaving every team more AI-native than it found them. 00 — Enablement arrow_forward 01 — Scope arrow_forward 02 — Build arrow_forward 03 — Sustain arrow_forward Foundations · Runs across all phases\n\nAI-fluent teams, before any build.\nSet the foundations so non-technical staff, engineers, and leadership all speak the same AI language. Most of our work lands harder when these are already in place. Foundations: AI policy, governance, and responsible-use guardrails Non-technical teams: 10x productivity, AI-native ways of working Engineering teams: best-in-class full-stack AI engineers Leadership: PE partners, operating partners, C-level\n\nThree things. Rarely found together.\nMost AI firms are technical but don’t understand value creation. Most consultants understand value creation but lack AI expertise. We bring both, with proven playbooks for AI deployments tied to value creation.", + "heading": "AI-fluent teams, before any build.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-9b92f5b623dd790f20e625db", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-37716df6766b4ec5a1e99dd4", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-920d55007c0e7dfcc6163188", + "text": "Community keyboard_arrow_down Mentorship Discord" + }, + { + "span_id": "span-acc096af8313e4d6eddb6229", + "text": "Careers — we're hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-cddc259bb00f81f5eedfe5c4", + "text": "Explore courses Talk to us" + }, + { + "span_id": "span-b3f0fbe6dd2476e390c3ec5e", + "text": "Proprietary AI playbooks , stress-tested in PE-backed environments to move real EBITDA ." + }, + { + "span_id": "span-6da33aa52d7a7678f93f0bb0", + "text": "Request a briefing Request a briefing See our work See our work" + }, + { + "span_id": "span-b7c1d01563fe67295523e74d", + "text": "Built by PE investors," + }, + { + "span_id": "span-9a61a610bdd9de0ca9f8f079", + "text": "operators, AI engineers" + }, + { + "span_id": "span-1ad411c0cdc203bff776ea0e", + "text": "We offer enablement + 3 engagement types, sold standalone or in phases, leaving every team more AI-native than it found them." + }, + { + "span_id": "span-fdf9ff49c7cdac638b293545", + "text": "00 — Enablement arrow_forward 01 — Scope arrow_forward 02 — Build arrow_forward 03 — Sustain arrow_forward" + }, + { + "span_id": "span-0ac8f88e61b0aa6af0577bfb", + "text": "Foundations · Runs across all phases" + }, + { + "span_id": "span-0988c1c974f6f7fec8c81194", + "text": "Set the foundations so non-technical staff, engineers, and leadership all speak the same AI language." + }, + { + "span_id": "span-9e50c8d87d2abb8b2fea7074", + "text": "Most of our work lands harder when these are already in place." + }, + { + "span_id": "span-021a7be15b2ddbc104755422", + "text": "Foundations: AI policy, governance, and responsible-use guardrails" + }, + { + "span_id": "span-706f93405304a70bad351790", + "text": "Non-technical teams: 10x productivity, AI-native ways of working" + }, + { + "span_id": "span-6db10174da10c44ec570f4c4", + "text": "Engineering teams: best-in-class full-stack AI engineers" + }, + { + "span_id": "span-233c3d4847e589074cfb410e", + "text": "Leadership: PE partners, operating partners, C-level" + }, + { + "span_id": "span-63a902f30440c91cd3ab6d44", + "text": "Most AI firms are technical but don’t understand value creation." + }, + { + "span_id": "span-afc7aa95cb7c58c08eabf2f5", + "text": "Most consultants understand value creation but lack AI expertise." + }, + { + "span_id": "span-4b27e4f01c3410f497f0cc2c", + "text": "We bring both, with proven playbooks for AI deployments tied to value creation." + } + ] + }, + { + "chunk_id": "tai-59d5268c7edbb6d75fef6c4f", + "text": "Finance expertise.\nOur co-founders come from J.P. Morgan, where they led trading, credit research, and asset management, and from Aurelius, a leading buy-and-build PE fund. We model EBITDA levers the way a finance team would, not the way an AI lab would.\n\nAI expertise.\nOur co-founders are recognised AI thought leaders with 200k+ followers and a best-selling book, Building LLMs for Production, used as a reference inside enterprise AI teams. 10,000+ copies sold.\n\nDeployment expertise.\nOur Towards AI Deployment practice ships custom systems through engineering pods, with strategists and leaders from Palantir, BCG, UBS, J.P. Morgan, Aurelius Growth Capital, and MILA. We know how AI lands in real organisations because we have been the people landing it.", + "heading": "Deployment expertise.", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-5ece3be43fee70d14ebed6aa", + "text": "Our co-founders come from J.P." + }, + { + "span_id": "span-00eda98f4292c3f4541d31c1", + "text": "Morgan, where they led trading, credit research, and asset management, and from Aurelius, a leading buy-and-build PE fund." + }, + { + "span_id": "span-0bfdbe7259d39bb65fdeaa7f", + "text": "We model EBITDA levers the way a finance team would, not the way an AI lab would." + }, + { + "span_id": "span-b604ab6a16f54f9437c85e6c", + "text": "Our co-founders are recognised AI thought leaders with 200k+ followers and a best-selling book, Building LLMs for Production, used as a reference inside enterprise AI teams." + }, + { + "span_id": "span-7a630ffa1166a8bab4249297", + "text": "10,000+ copies sold." + }, + { + "span_id": "span-e453e164433f7762a4332e41", + "text": "Our Towards AI Deployment practice ships custom systems through engineering pods, with strategists and leaders from Palantir, BCG, UBS, J.P." + }, + { + "span_id": "span-30a02bdbb20e9ad5a50ceef5", + "text": "Morgan, Aurelius Growth Capital, and MILA." + }, + { + "span_id": "span-31efc72f55151dd4df2c2cd6", + "text": "We know how AI lands in real organisations because we have been the people landing it." + } + ] + }, + { + "chunk_id": "tai-ba9b8a5e43066c5570d0b75e", + "text": "Recent engagements .\nSaved hours are capacity, not EBITDA. We count impact only when the operating plan changes. 🇺🇸 Leading PE Fund HVAC services buy & build · USA Problem statement Workers logged material costs per job on paper. Backoffice manually re-typed from paper into Excel. Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails. What we built AI pipeline that takes PDFs of paper receipts Automatic extraction & calculation of material costs per Job Real-time consolidation & job-level reporting Audit-ready monthly reports Outcome Hours of finance team time saved per week. Audit-ready reports available instantly, not at month-end. 🇬🇧 UK IT Services Group IT services buy & build · UK Problem statement Multiple team members spent hours daily scanning fragmented tender sites across many geographies. Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most. What we built Automated scraping across 5+ tender platforms AI ranking by relevance to the group’s strengths AI categorization of likely win-rate Daily report of newly-relevant opportunities Outcome Many hours of analyst time recovered per week. Pipeline focused on the tenders the group is structurally most likely to win. 🇫🇷 Europol Law enforcement agency · France Problem statement Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines. What we built Multi-day intensive bootcamp on LLM pipelines Tailored to law-enforcement workflows Secure deployment for sensitive intelligence work Hands-on labs for context engineering Outcome Special agents trained and ready to deploy AI tools across operational", + "heading": "Recent engagements .", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-7f23dbeb229f4717c1812e93", + "text": "Saved hours are capacity, not EBITDA." + }, + { + "span_id": "span-47f3a598dc5da7aad9d9574c", + "text": "We count impact only when the operating plan changes." + }, + { + "span_id": "span-8f190f0ff3ab4f3155e429f0", + "text": "🇺🇸 Leading PE Fund" + }, + { + "span_id": "span-03cd95d745507b5b2d3046c2", + "text": "HVAC services buy & build · USA" + }, + { + "span_id": "span-ae58df1ce5951a0742a6d34d", + "text": "Problem statement" + }, + { + "span_id": "span-5416960e388913e0b4521457", + "text": "Workers logged material costs per job on paper." + }, + { + "span_id": "span-2683c85fcb9a557bffd63f97", + "text": "Backoffice manually re-typed from paper into Excel." + }, + { + "span_id": "span-17be48c682927663fb69c5ea", + "text": "Backlog piled up increasing Sales Outstanding, while monthly cost reporting was unstructured, leading to weaker audit trails." + }, + { + "span_id": "span-ee3bd49eb855673cc1817232", + "text": "What we built" + }, + { + "span_id": "span-d5d9dcc1f4b1fece61d4cb7a", + "text": "AI pipeline that takes PDFs of paper receipts" + }, + { + "span_id": "span-f73d248e5a5ba1b3358e4cff", + "text": "Automatic extraction & calculation of material costs per Job" + }, + { + "span_id": "span-2d4a9df7088fe08ca4d2eda4", + "text": "Real-time consolidation & job-level reporting" + }, + { + "span_id": "span-2c216aa1fad82d9fab623d68", + "text": "Audit-ready monthly reports" + }, + { + "span_id": "span-3f9536342bf7edccdb9f79ea", + "text": "Hours of finance team time saved per week." + }, + { + "span_id": "span-203c06e98266d44d5b773a84", + "text": "Audit-ready reports available instantly, not at month-end." + }, + { + "span_id": "span-ec1c285557d6557dffd53345", + "text": "🇬🇧 UK IT Services Group" + }, + { + "span_id": "span-e20659b52635664aff5f340d", + "text": "IT services buy & build · UK" + }, + { + "span_id": "span-d6559356da36a40351be4f73", + "text": "Multiple team members spent hours daily scanning fragmented tender sites across many geographies." + }, + { + "span_id": "span-34d88ce88877bfbf4d81dabd", + "text": "Easy to miss high-value tenders and impossible to quickly prioritise the ones worth pursuing most." + }, + { + "span_id": "span-c515592b85dcfd6560eb9966", + "text": "Automated scraping across 5+ tender platforms" + }, + { + "span_id": "span-8570665c113efe20d1bf87cf", + "text": "AI ranking by relevance to the group’s strengths" + }, + { + "span_id": "span-cf65d1aff6689b624b021eb1", + "text": "AI categorization of likely win-rate" + }, + { + "span_id": "span-fb88a230970695fe016e57ca", + "text": "Daily report of newly-relevant opportunities" + }, + { + "span_id": "span-b57c3e8de4b167b5dd39ed8f", + "text": "Many hours of analyst time recovered per week." + }, + { + "span_id": "span-b9c81d00ace8b88e30d6e61c", + "text": "Pipeline focused on the tenders the group is structurally most likely to win." + }, + { + "span_id": "span-eb91f9323395faf190608ff6", + "text": "Law enforcement agency · France" + }, + { + "span_id": "span-367386b089ae38f0c153916a", + "text": "Special agents needed to apply state-of-the-art AI tools to mission-critical intelligence workflows, but lacked technical confidence with modern LLM pipelines." + }, + { + "span_id": "span-e69b8b85227caa27d1a9e8cf", + "text": "Multi-day intensive bootcamp on LLM pipelines" + }, + { + "span_id": "span-5e7b7f983f47662ad7cfe0d3", + "text": "Tailored to law-enforcement workflows" + }, + { + "span_id": "span-38147719b60001ee733e2207", + "text": "Secure deployment for sensitive intelligence work" + }, + { + "span_id": "span-7145ff3dd50683b7e7796a47", + "text": "Hands-on labs for context engineering" + } + ] + }, + { + "chunk_id": "tai-3f2073897ad03d147f418016", + "text": "Recent engagements .\nintelligence workflows. 🇪🇺 Nviya-Prime Agentic financial intelligence · Europe The brief Build an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when, for an agentic B2B financial-technology platform founded by a former banking partner. What we delivered Agentic platform combining filings, live news, risk data and future-event signals Prioritised, product-specific pitch actions for bankers Around 800 prompts and one million words of research per company Compliance-led deployment across Gemini, OpenAI, Claude and Grok Outcome Banker reasoning reproduced at scale across client-cloud, API-only and locally hosted configurations, in live demonstrations with global banks.\n\nEvery portfolio company has an EBITDA lever . Let’s find yours.\nA 30-minute call with the Towards AI leadership team. We’ll discuss the value-creation priorities on your desk, where AI can move the number, and the shape of an engagement. Request a briefing arrow_forward Email us Reply within one business day.", + "heading": "Recent engagements .", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-087fa66e048bc2f8b101cb6e", + "text": "🇪🇺 Nviya-Prime" + }, + { + "span_id": "span-101dc436a77d2b64ff6d6c18", + "text": "Agentic financial intelligence · Europe" + }, + { + "span_id": "span-135b8ee88f965b5f57c60e41", + "text": "The brief" + }, + { + "span_id": "span-df9bbe3d5f801428f10ee090", + "text": "Build an agentic engine that reproduces banker reasoning to identify what to pitch, to whom, and when, for an agentic B2B financial-technology platform founded by a former banking partner." + }, + { + "span_id": "span-d92c3b5b99ae321c2bf4c1e3", + "text": "What we delivered" + }, + { + "span_id": "span-57a8e72324a201d3a804e91d", + "text": "Agentic platform combining filings, live news, risk data and future-event signals" + }, + { + "span_id": "span-4321465185c4fcd10f6dadbb", + "text": "Prioritised, product-specific pitch actions for bankers" + }, + { + "span_id": "span-adb452fb472dceecf25ea442", + "text": "Around 800 prompts and one million words of research per company" + }, + { + "span_id": "span-ccdccd0b5247fbcd1b07f4b9", + "text": "Compliance-led deployment across Gemini, OpenAI, Claude and Grok" + }, + { + "span_id": "span-6c323be4bf5b41e08c3891c4", + "text": "Banker reasoning reproduced at scale across client-cloud, API-only and locally hosted configurations, in live demonstrations with global banks." + }, + { + "span_id": "span-e0c2125c25f1889fe6226184", + "text": "A 30-minute call with the Towards AI leadership team." + }, + { + "span_id": "span-3c17425b3cdd1d4f82085cab", + "text": "We’ll discuss the value-creation priorities on your desk, where AI can move the number, and the shape of an engagement." + }, + { + "span_id": "span-750d9ab390c3c62b6b967ed0", + "text": "Request a briefing arrow_forward Email us" + }, + { + "span_id": "span-888738504dd2577d6409498f", + "text": "Reply within one business day." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "d0f862be0fcfca578bac771459d173c1eaae769ee17033e6fd5ce2e68b9275d5", + "content_hash": "d0f862be0fcfca578bac771459d173c1eaae769ee17033e6fd5ce2e68b9275d5", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/valuecreation/" + ], + "evidence_hash": "407821ccc2040f39dfac3b98a6ad2f6d3480944dd1c00598a5676619b20e7628" + }, + { + "discovered_url": "https://towardsai.com/valuecreation/careers/", + "url": "https://towardsai.com/valuecreation/careers/", + "canonical_url": "https://towardsai.com/valuecreation/careers/", + "host": "towardsai.com", + "path": "/valuecreation/careers", + "kind": "career", + "offer_id": null, + "entity_id": "page:towardsai.com/valuecreation/careers", + "title": "Careers — Towards AI", + "review_title": "", + "reviewed_summary": "", + "meta_description": "Join Towards AI. We build production AI inside PE-backed mid-market companies. Open roles for Deployment Strategists and AI Engineers.", + "headings": [ + "Build AI tools that get used.", + "The real AI opportunity is in the mid-market, not the Fortune 500.", + "AI has the biggest impact in SMEs.", + "SMEs share the same bottlenecks.", + "Adoption is structural, not nice-to-have.", + "You’ll work at the frontier — shaping new playbooks with the latest technology.", + "Deployment Strategist.", + "AI Engineer.", + "Why this team.", + "Education and deployment, one firm", + "The people", + "Latest tools, always", + "Proven playbooks", + "Five conversations.", + "Initial conversation", + "Technical assessment", + "Project presentation", + "Cultural fit", + "The offer", + "Frequently asked questions.", + "Ready to build?" + ], + "text": "We’re hiring\n\nBuild AI tools that get used.\nI'm a Deployment Strategist Learn the proven PE playbooks that actually get implemented. arrow_forward I'm a full-stack developer We turn full-stack devs into tier-1 AI engineers who own a real system end to end, taught by the people behind a 500k-practitioner learning platform. arrow_forward Discover Towards AI arrow_outward Join a team of engineers, investors, operators, and ex-founders The opportunity ahead\n\nThe real AI opportunity is in the mid-market, not the Fortune 500.\ntrending_up SME economics\n\nAI has the biggest impact in SMEs.\nLarge enterprises run at high efficiency with ‘fortified’ IT and slow procurement. SMEs have obvious bottlenecks and short feedback loops. The marginal impact of every AI deployment is not only bigger, you see it in the P&L. replay Repeatable playbooks\n\nSMEs share the same bottlenecks.\nProcurement, document workflows, sales pipeline triage, customer service. We’ve built playbooks for each. Every new engagement compounds on the last. sync_alt PE as adoption driver\n\nAdoption is structural, not nice-to-have.\nInside PE-backed portcos, weekly operating cadences, focused KPIs, and CEO incentives ensure tools we build actually get used. The work ships and gets adopted by design. auto_awesome No legacy drag\n\nYou’ll work at the frontier — shaping new playbooks with the latest technology.\nWe don’t maintain a long-lived product platform. Each engagement is a fresh deployment, so you’re using the best AI tooling and defining the most effective use case every time. Open role\n\nDeployment Strategist.\nTranslate bottlenecks and workflow pain points into AI-enabled workflows. Own client engagements at PE-backed portcos end-to-end. Who you are Consistently performed at a high level in tier-1 institutions You take pride in measurable impact in real companies, not slide decks Comfortable owning client relationships and engagement outcomes 2+ years experience What you'll do Own deployment engagements at PE-backed portcos Map workflows, identify high-impact automations, scope solutions Run delivery from kickoff to live system Stakeholder management across PE partners, portco C-level, and technical teams What you'll learn PE playbooks that actually get implemented Work alongside top-tier PE partners and operating partners at leading funds Use the latest AI tools to stay at the forefront of the field Proven workflows across procurement, sales, finance, and document automation Open role\n\nAI Engineer.\nBuild production AI with the latest tools. We turn full-stack devs into tier-1 AI engineers through our best-in-class learning platform. Who you are 2+ years as a full-stack developer You work fluently with modern dev tools (TypeScript, Python, React, Postgres, Docker, Git) You want to learn how to use AI tools most effectively You regularly use AI coding tools (Cursor, Claude Code, Copilot, or similar) What you'll do Build production AI systems inside PE-backed companies RAG pipelines, agents, document automation, custom tooling Pick the right model, framework, and tradeoffs each time Own deployments end-to-end What you'll learn Direct upskilling from the authors of Building LLMs for Production (10k+ copies sold) The frontier of what works in production AI Internal playbooks across procurement, sales, finance, document workflows How to ship AI inside real businesses, rarer than it sounds What you get\n\nWhy this team.\nGlobally distributed , headquartered in London. verified\n\nEducation and deployment, one firm\n200k+ newsletter readers. 100k+ community members. 500k+ practitioners taught since 2019. Our book is used inside enterprise AI teams; our courses run on O’Reilly. The path runs community → Learning → Deployment pod: you join the pods that own real systems end to end. workspaces\n\nThe people\nOur team has a breadth of experience across leading companies, having seen the PE value-creation and AI tool-building process end-to-end. auto_awesome\n\nLatest tools, always\nNo legacy platform to maintain. Every project uses the best AI tooling available at the time of building. menu_book\n\nProven playbooks\nWe've shipped AI inside multiple PE portcos. We know what moves the needle and what doesn't. How recruitment works\n\nFive conversations.\nWe read every application individually, which takes time. We also receive more applications than we can interview. From first call to offer typically takes 2–4 weeks. We keep candidates updated at every stage. 01\n\nInitial conversation\n45 min A get-to-know-you call. We’ll walk through your experience using the STAR framework, what you’ve shipped, and what motivates you. 02\n\nTechnical assessment\n1 hour For Engineers: a live coding and AI architecture session. For Strategists: a case study on how you’d approach a real PE-backed deployment. 03\n\nProject presentation\n1 hour In-person or video. 15 min : you walk through a recent project you’ve owned. 45 min : Q&A with senior leaders. For Strategists we look for presentation skills, executive presence, and how you think. For Engineers we focus on technical capability and how you think. 04\n\nCultural fit\nSenior team call A relaxed conversation with senior members of the team. Really just a chance to make sure this feels like the right home for you, and that we’d build well together. 05\n\nThe offer\nIf the fit feels right on both sides, we’d be delighted to welcome you to the team. FAQs\n\nFrequently asked questions.\nWe try our best to answer every question within a few days. Unfortunately we receive more emails than we have time to review, so please have a look below first; we’ve covered the most common questions. If yours isn’t answered here, do send it through afterwards. Does Towards AI use AI to assess job applicants? expand_more No, we don’t. Every application is read by a real person on our team. We think that’s the right thing to do, especially for an AI company. The work matters too much to outsource the first impression of you to a model. When should I expect to hear back after I apply? expand_more We aim to reply to everyone within a week, one way or the other. Sometimes it takes a little longer when we’re working through a wave of applications. If you haven’t heard from us by then, please don’t hesitate to nudge us. Can I apply for more than one open role? expand_more We currently have two open roles: AI Engineer and Deployment Strategist. Pick the one that feels most like you, and apply there. If we think the other role might suit you better, we’ll bring it up during the process. Does Towards AI give application or interview feedback? expand_more If you make it to the interview stage and things don’t work out, we’ll do our best to share useful thoughts within a few days. We genuinely wish we could give feedback to everyone who applies, but the volume makes that impossible. We’d rather be honest about that than promise something we can’t deliver. Can I reapply if I was unsuccessful the first time? expand_more Yes, though we’d gently suggest waiting at least a year before reapplying for the same role. If we come across a different opening that feels like a better fit for your background, we’ll reach out. Following our team on LinkedIn and X is the best way to catch new roles as they open. Does Towards AI pay for travel costs for on-site interviews? expand_more Most of our interviews happen over Google Meet or Zoom, so travel usually isn’t needed. On the rare occasion we invite you to meet in person, we’re happy to cover reasonable travel and accommodation. Just let us know what you need. I’m a recruiter. Can I help Towards AI find more talent? expand_more Thanks for the offer, and we mean that genuinely. For now we’ve built our own trusted networks for hiring, so we rarely need external support. If that changes, we work with a small list of preferred partners that we review regularly. How to apply\n\nReady to build?\nSend your LinkedIn, CV, and a short note on why this role interests you. We read every email and reply within a week. Email denis@towardsai.net arrow_forward Competitive comp + equity.", + "links": [ + { + "text": "Discover Towards AI arrow_outward", + "url": "https://towardsai.com/" + }, + { + "text": "Globally distributed", + "url": "https://towardsai.com/enterpriseenablement/" + } + ], + "chunks": [ + { + "chunk_id": "tai-dd23a5d27c94b4bc65865609", + "text": "We’re hiring\n\nBuild AI tools that get used.\nI'm a Deployment Strategist Learn the proven PE playbooks that actually get implemented. arrow_forward I'm a full-stack developer We turn full-stack devs into tier-1 AI engineers who own a real system end to end, taught by the people behind a 500k-practitioner learning platform. arrow_forward Discover Towards AI arrow_outward Join a team of engineers, investors, operators, and ex-founders The opportunity ahead\n\nThe real AI opportunity is in the mid-market, not the Fortune 500.\ntrending_up SME economics\n\nAI has the biggest impact in SMEs.\nLarge enterprises run at high efficiency with ‘fortified’ IT and slow procurement. SMEs have obvious bottlenecks and short feedback loops. The marginal impact of every AI deployment is not only bigger, you see it in the P&L. replay Repeatable playbooks\n\nSMEs share the same bottlenecks.\nProcurement, document workflows, sales pipeline triage, customer service. We’ve built playbooks for each. Every new engagement compounds on the last. sync_alt PE as adoption driver\n\nAdoption is structural, not nice-to-have.\nInside PE-backed portcos, weekly operating cadences, focused KPIs, and CEO incentives ensure tools we build actually get used. The work ships and gets adopted by design. auto_awesome No legacy drag\n\nYou’ll work at the frontier — shaping new playbooks with the latest technology.\nWe don’t maintain a long-lived product platform. Each engagement is a fresh deployment, so you’re using the best AI tooling and defining the most effective use case every time. Open role", + "heading": "Build AI tools that get used.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-278f91325a02bfbbdd8debc5", + "text": "We’re hiring" + }, + { + "span_id": "span-ccebce0cd8042969ad11ad4b", + "text": "I'm a Deployment Strategist" + }, + { + "span_id": "span-8401705c2d43a56e7ba70a50", + "text": "Learn the proven PE playbooks that actually get implemented." + }, + { + "span_id": "span-201dae04a6e8dfde7550659a", + "text": "arrow_forward I'm a full-stack developer" + }, + { + "span_id": "span-8fab5510f682b7f766af13ce", + "text": "We turn full-stack devs into tier-1 AI engineers who own a real system end to end, taught by the people behind a 500k-practitioner learning platform." + }, + { + "span_id": "span-4da8a5c51d9f9c870948241c", + "text": "arrow_forward" + }, + { + "span_id": "span-e86bd0885d27a6dd712448de", + "text": "Discover Towards AI arrow_outward" + }, + { + "span_id": "span-4c62c2cd86037525b2ad6b72", + "text": "Join a team of engineers, investors, operators, and ex-founders" + }, + { + "span_id": "span-b7fd352625c9a9b9eb58a4ed", + "text": "The opportunity ahead" + }, + { + "span_id": "span-21e2a38391587ed6cfd60cb5", + "text": "The real AI opportunity is in the mid-market, not the Fortune 500." + }, + { + "span_id": "span-7508be350c80c1e612d02470", + "text": "trending_up" + }, + { + "span_id": "span-b945cbe5bf5c441a7453a335", + "text": "SME economics" + }, + { + "span_id": "span-9cb7c55ecc1f4d37dd7d99f5", + "text": "Large enterprises run at high efficiency with ‘fortified’ IT and slow procurement." + }, + { + "span_id": "span-83b9e58e367c7b1d56cc7a5e", + "text": "SMEs have obvious bottlenecks and short feedback loops." + }, + { + "span_id": "span-8ce3d5ebac1adae7e185cb50", + "text": "The marginal impact of every AI deployment is not only bigger, you see it in the P&L." + }, + { + "span_id": "span-05faef17e81d7720779964b3", + "text": "Repeatable playbooks" + }, + { + "span_id": "span-cd9a8675bc99e1a6bda19ed4", + "text": "Procurement, document workflows, sales pipeline triage, customer service." + }, + { + "span_id": "span-e170750ffcfee62a51275486", + "text": "We’ve built playbooks for each." + }, + { + "span_id": "span-08159b3bc36a7c1a5d131ff3", + "text": "Every new engagement compounds on the last." + }, + { + "span_id": "span-0ec1a2885dd0b10a1e49761a", + "text": "sync_alt" + }, + { + "span_id": "span-acada69310dfe87baedcc204", + "text": "PE as adoption driver" + }, + { + "span_id": "span-3a2ab03dbd9f097b9c670910", + "text": "Inside PE-backed portcos, weekly operating cadences, focused KPIs, and CEO incentives ensure tools we build actually get used." + }, + { + "span_id": "span-5f7990e68349efc5912fe8df", + "text": "The work ships and gets adopted by design." + }, + { + "span_id": "span-79a0ce136488c0426b88a435", + "text": "auto_awesome" + }, + { + "span_id": "span-ed07a857eed0e0e71e4c664b", + "text": "No legacy drag" + }, + { + "span_id": "span-be169fd7bcf12af7d4512a0c", + "text": "We don’t maintain a long-lived product platform." + }, + { + "span_id": "span-b1145992fcdf2803392982f2", + "text": "Each engagement is a fresh deployment, so you’re using the best AI tooling and defining the most effective use case every time." + }, + { + "span_id": "span-60e34ad12d337c43a7bdd40c", + "text": "Open role" + } + ] + }, + { + "chunk_id": "tai-ee989199f4f476bbab8ddf63", + "text": "Deployment Strategist.\nTranslate bottlenecks and workflow pain points into AI-enabled workflows. Own client engagements at PE-backed portcos end-to-end. Who you are Consistently performed at a high level in tier-1 institutions You take pride in measurable impact in real companies, not slide decks Comfortable owning client relationships and engagement outcomes 2+ years experience What you'll do Own deployment engagements at PE-backed portcos Map workflows, identify high-impact automations, scope solutions Run delivery from kickoff to live system Stakeholder management across PE partners, portco C-level, and technical teams What you'll learn PE playbooks that actually get implemented Work alongside top-tier PE partners and operating partners at leading funds Use the latest AI tools to stay at the forefront of the field Proven workflows across procurement, sales, finance, and document automation Open role", + "heading": "Deployment Strategist.", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-ae5e4eebdfc177b5069e1b28", + "text": "Translate bottlenecks and workflow pain points into AI-enabled workflows." + }, + { + "span_id": "span-f4222b824a870a5fcbfa0106", + "text": "Own client engagements at PE-backed portcos end-to-end." + }, + { + "span_id": "span-4d8ebd93da93a1fb60c546da", + "text": "Who you are" + }, + { + "span_id": "span-0c53c0f32dc4f326179ffeb2", + "text": "Consistently performed at a high level in tier-1 institutions" + }, + { + "span_id": "span-21e04365cc8acdb30aabf623", + "text": "You take pride in measurable impact in real companies, not slide decks" + }, + { + "span_id": "span-61cb847b72fcc3f5f3864d1f", + "text": "Comfortable owning client relationships and engagement outcomes" + }, + { + "span_id": "span-e80d25e77d47772028adcf39", + "text": "2+ years experience" + }, + { + "span_id": "span-25c817805f39d8a965c39173", + "text": "What you'll do" + }, + { + "span_id": "span-cfd2e79c63c13ec8d773b238", + "text": "Own deployment engagements at PE-backed portcos" + }, + { + "span_id": "span-63838595e70187967b299cd7", + "text": "Map workflows, identify high-impact automations, scope solutions" + }, + { + "span_id": "span-d5c25c4380d96e9b6a6ca5a2", + "text": "Run delivery from kickoff to live system" + }, + { + "span_id": "span-13a8048475f39321ee3c6e65", + "text": "Stakeholder management across PE partners, portco C-level, and technical teams" + }, + { + "span_id": "span-7608da4614b6ffd532a5a64d", + "text": "What you'll learn" + }, + { + "span_id": "span-074dfa2fc9ae8cc5c0a2a2b1", + "text": "PE playbooks that actually get implemented" + }, + { + "span_id": "span-bad9b5a00bd95c5938aacb77", + "text": "Work alongside top-tier PE partners and operating partners at leading funds" + }, + { + "span_id": "span-7c413ed01ed04a695a6f15c0", + "text": "Use the latest AI tools to stay at the forefront of the field" + }, + { + "span_id": "span-b1d0c76e749075714bc4b173", + "text": "Proven workflows across procurement, sales, finance, and document automation" + }, + { + "span_id": "span-4084845173d1185d03231eb6", + "text": "Open role" + } + ] + }, + { + "chunk_id": "tai-fe6dd482d169d44fa6c068a4", + "text": "AI Engineer.\nBuild production AI with the latest tools. We turn full-stack devs into tier-1 AI engineers through our best-in-class learning platform. Who you are 2+ years as a full-stack developer You work fluently with modern dev tools (TypeScript, Python, React, Postgres, Docker, Git) You want to learn how to use AI tools most effectively You regularly use AI coding tools (Cursor, Claude Code, Copilot, or similar) What you'll do Build production AI systems inside PE-backed companies RAG pipelines, agents, document automation, custom tooling Pick the right model, framework, and tradeoffs each time Own deployments end-to-end What you'll learn Direct upskilling from the authors of Building LLMs for Production (10k+ copies sold) The frontier of what works in production AI Internal playbooks across procurement, sales, finance, document workflows How to ship AI inside real businesses, rarer than it sounds What you get\n\nWhy this team.\nGlobally distributed , headquartered in London. verified\n\nEducation and deployment, one firm\n200k+ newsletter readers. 100k+ community members. 500k+ practitioners taught since 2019. Our book is used inside enterprise AI teams; our courses run on O’Reilly. The path runs community → Learning → Deployment pod: you join the pods that own real systems end to end. workspaces\n\nThe people\nOur team has a breadth of experience across leading companies, having seen the PE value-creation and AI tool-building process end-to-end. auto_awesome\n\nLatest tools, always\nNo legacy platform to maintain. Every project uses the best AI tooling available at the time of building. menu_book", + "heading": "AI Engineer.", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-d63b9b73cdc55d35313b1b7b", + "text": "Build production AI with the latest tools." + }, + { + "span_id": "span-06a2d472b3b3e57c2c5d2d7c", + "text": "We turn full-stack devs into tier-1 AI engineers through our best-in-class learning platform." + }, + { + "span_id": "span-ea7772b5258a7df56ce0e3a3", + "text": "Who you are" + }, + { + "span_id": "span-f0b1a67efc16fd4e8026a3ff", + "text": "2+ years as a full-stack developer" + }, + { + "span_id": "span-0fc3d55ce79078ad9f913e46", + "text": "You work fluently with modern dev tools (TypeScript, Python, React, Postgres, Docker, Git)" + }, + { + "span_id": "span-98b1a08cb19e588ec2a13c6e", + "text": "You want to learn how to use AI tools most effectively" + }, + { + "span_id": "span-8373bb95599b7b6356d227e6", + "text": "You regularly use AI coding tools (Cursor, Claude Code, Copilot, or similar)" + }, + { + "span_id": "span-192109e6107d30db1b1612c8", + "text": "What you'll do" + }, + { + "span_id": "span-7d548a95af78af60c7aad9fb", + "text": "Build production AI systems inside PE-backed companies" + }, + { + "span_id": "span-483365935b985dd9e23a39b6", + "text": "RAG pipelines, agents, document automation, custom tooling" + }, + { + "span_id": "span-1797a0868444428e806673db", + "text": "Pick the right model, framework, and tradeoffs each time" + }, + { + "span_id": "span-8c0c14edcff9252513e7fa1a", + "text": "Own deployments end-to-end" + }, + { + "span_id": "span-7251e97a8b2f735f42ede737", + "text": "What you'll learn" + }, + { + "span_id": "span-ff896e1cbac4bf0ad438f401", + "text": "Direct upskilling from the authors of Building LLMs for Production (10k+ copies sold)" + }, + { + "span_id": "span-eea8c9955255c34a2a388198", + "text": "The frontier of what works in production AI" + }, + { + "span_id": "span-1b383d7e9496c217a4529bc7", + "text": "Internal playbooks across procurement, sales, finance, document workflows" + }, + { + "span_id": "span-62d87cfa160edbc7200f7298", + "text": "How to ship AI inside real businesses, rarer than it sounds" + }, + { + "span_id": "span-58f9e21282d97b02eafbc2de", + "text": "What you get" + }, + { + "span_id": "span-14baa48a3f26014b0f76a398", + "text": "Globally distributed , headquartered in London." + }, + { + "span_id": "span-76e1cf66d7998102c3c31b23", + "text": "200k+ newsletter readers." + }, + { + "span_id": "span-fbaf3993ab2564ea92cfa5e4", + "text": "100k+ community members." + }, + { + "span_id": "span-4ddc2c020396eae83221e6a9", + "text": "500k+ practitioners taught since 2019." + }, + { + "span_id": "span-64c11427ba13e7e2ae88f25e", + "text": "Our book is used inside enterprise AI teams; our courses run on O’Reilly." + }, + { + "span_id": "span-22eabac916be8f3701f77092", + "text": "The path runs community → Learning → Deployment pod: you join the pods that own real systems end to end." + }, + { + "span_id": "span-fec097c4c392d28844d9fc7e", + "text": "Our team has a breadth of experience across leading companies, having seen the PE value-creation and AI tool-building process end-to-end." + }, + { + "span_id": "span-a68e11b99b907d73a01156ee", + "text": "auto_awesome" + }, + { + "span_id": "span-aff540243804cdb9a73c4b74", + "text": "No legacy platform to maintain." + }, + { + "span_id": "span-5818806886c2b3c5c4a2a9a5", + "text": "Every project uses the best AI tooling available at the time of building." + }, + { + "span_id": "span-ca60e7f528ce2fbe91387112", + "text": "menu_book" + } + ] + }, + { + "chunk_id": "tai-46ec6da138270071e65a1c68", + "text": "Proven playbooks\nWe've shipped AI inside multiple PE portcos. We know what moves the needle and what doesn't. How recruitment works\n\nFive conversations.\nWe read every application individually, which takes time. We also receive more applications than we can interview. From first call to offer typically takes 2–4 weeks. We keep candidates updated at every stage. 01\n\nInitial conversation\n45 min A get-to-know-you call. We’ll walk through your experience using the STAR framework, what you’ve shipped, and what motivates you. 02\n\nTechnical assessment\n1 hour For Engineers: a live coding and AI architecture session. For Strategists: a case study on how you’d approach a real PE-backed deployment. 03\n\nProject presentation\n1 hour In-person or video. 15 min : you walk through a recent project you’ve owned. 45 min : Q&A with senior leaders. For Strategists we look for presentation skills, executive presence, and how you think. For Engineers we focus on technical capability and how you think. 04\n\nCultural fit\nSenior team call A relaxed conversation with senior members of the team. Really just a chance to make sure this feels like the right home for you, and that we’d build well together. 05\n\nThe offer\nIf the fit feels right on both sides, we’d be delighted to welcome you to the team. FAQs", + "heading": "Project presentation", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-4e112fb9808c1ab74a5289f4", + "text": "We've shipped AI inside multiple PE portcos." + }, + { + "span_id": "span-35e3236e3eeeef8cef8502fd", + "text": "We know what moves the needle and what doesn't." + }, + { + "span_id": "span-140be3f4543954b852cbb1b6", + "text": "How recruitment works" + }, + { + "span_id": "span-05af48dfc8b9fb704d986650", + "text": "We read every application individually, which takes time." + }, + { + "span_id": "span-5cace0f3052496b267df2d76", + "text": "We also receive more applications than we can interview." + }, + { + "span_id": "span-0f9950c38f08df57a7675c17", + "text": "From first call to offer typically takes 2–4 weeks." + }, + { + "span_id": "span-32dd0f8005154dada26bc231", + "text": "We keep candidates updated at every stage." + }, + { + "span_id": "span-6f49b802abf729566635a43a", + "text": "45 min" + }, + { + "span_id": "span-2c23b32ebfb2414b204babea", + "text": "A get-to-know-you call." + }, + { + "span_id": "span-27d8b77308785123be0ff273", + "text": "We’ll walk through your experience using the STAR framework, what you’ve shipped, and what motivates you." + }, + { + "span_id": "span-54499a5067a7b8295fbd1eba", + "text": "1 hour" + }, + { + "span_id": "span-6cf3c9eed9ab204634ebf723", + "text": "For Engineers: a live coding and AI architecture session." + }, + { + "span_id": "span-6eb15844760eb2ad6c411241", + "text": "For Strategists: a case study on how you’d approach a real PE-backed deployment." + }, + { + "span_id": "span-14eddd41242076e2f997fbaf", + "text": "In-person or video." + }, + { + "span_id": "span-63231f590123c44ddfff67a4", + "text": "15 min : you walk through a recent project you’ve owned." + }, + { + "span_id": "span-add2fd9f02eab16e5c64d39f", + "text": "45 min : Q&A with senior leaders." + }, + { + "span_id": "span-45bc2fa218174f9a69fbb9a5", + "text": "For Strategists we look for presentation skills, executive presence, and how you think." + }, + { + "span_id": "span-40cf221ac0c2a6231755b088", + "text": "For Engineers we focus on technical capability and how you think." + }, + { + "span_id": "span-740819c1a1f9b7c1573b84a5", + "text": "Senior team call" + }, + { + "span_id": "span-961a845dd294f0a42725368a", + "text": "A relaxed conversation with senior members of the team." + }, + { + "span_id": "span-2c35aee75d47182452c073e2", + "text": "Really just a chance to make sure this feels like the right home for you, and that we’d build well together." + }, + { + "span_id": "span-55dd963cc78f21bdfbff9824", + "text": "If the fit feels right on both sides, we’d be delighted to welcome you to the team." + } + ] + }, + { + "chunk_id": "tai-769f6b84fee7fcc014a80a96", + "text": "Frequently asked questions.\nWe try our best to answer every question within a few days. Unfortunately we receive more emails than we have time to review, so please have a look below first; we’ve covered the most common questions. If yours isn’t answered here, do send it through afterwards. Does Towards AI use AI to assess job applicants? expand_more No, we don’t. Every application is read by a real person on our team. We think that’s the right thing to do, especially for an AI company. The work matters too much to outsource the first impression of you to a model. When should I expect to hear back after I apply? expand_more We aim to reply to everyone within a week, one way or the other. Sometimes it takes a little longer when we’re working through a wave of applications. If you haven’t heard from us by then, please don’t hesitate to nudge us. Can I apply for more than one open role? expand_more We currently have two open roles: AI Engineer and Deployment Strategist. Pick the one that feels most like you, and apply there. If we think the other role might suit you better, we’ll bring it up during the process. Does Towards AI give application or interview feedback? expand_more If you make it to the interview stage and things don’t work out, we’ll do our best to share useful thoughts within a few days. We genuinely wish we could give feedback to everyone who applies, but the volume makes that impossible. We’d rather be honest about that than promise something we can’t deliver. Can I reapply if I was unsuccessful the first time? expand_more Yes, though we’d gently suggest waiting at least a year before reapplying for the same role. If we come across a different opening that feels like a better fit for your background, we’ll reach out. Following our team on LinkedIn and X is", + "heading": "Frequently asked questions.", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-3a4317badcc4191460e18627", + "text": "We try our best to answer every question within a few days." + }, + { + "span_id": "span-1e0a85b014905414f6a0c03b", + "text": "Unfortunately we receive more emails than we have time to review, so please have a look below first; we’ve covered the most common questions." + }, + { + "span_id": "span-64cbae5ea6d5234548a7293d", + "text": "If yours isn’t answered here, do send it through afterwards." + }, + { + "span_id": "span-4a9fcfd33257d2755d057233", + "text": "Does Towards AI use AI to assess job applicants? expand_more No, we don’t." + }, + { + "span_id": "span-95a2cd608adecb0f8d81841e", + "text": "Every application is read by a real person on our team." + }, + { + "span_id": "span-8da304619410ee9bc44358ec", + "text": "We think that’s the right thing to do, especially for an AI company." + }, + { + "span_id": "span-035dedcd145470282e497c57", + "text": "The work matters too much to outsource the first impression of you to a model." + }, + { + "span_id": "span-1eccfd8be196277b9cd2fa46", + "text": "When should I expect to hear back after I apply? expand_more We aim to reply to everyone within a week, one way or the other." + }, + { + "span_id": "span-67e95fa245ea8b4e55532e77", + "text": "Sometimes it takes a little longer when we’re working through a wave of applications." + }, + { + "span_id": "span-09203ba51bcbcf1435144dbb", + "text": "If you haven’t heard from us by then, please don’t hesitate to nudge us." + }, + { + "span_id": "span-a77713986f1d11331bd7c2d7", + "text": "Can I apply for more than one open role? expand_more We currently have two open roles: AI Engineer and Deployment Strategist." + }, + { + "span_id": "span-33d5fae50224468bd61c7a38", + "text": "Pick the one that feels most like you, and apply there." + }, + { + "span_id": "span-4777d1f3cbe513a91c601376", + "text": "If we think the other role might suit you better, we’ll bring it up during the process." + }, + { + "span_id": "span-e66d86fadd3f22fba4935e00", + "text": "Does Towards AI give application or interview feedback? expand_more If you make it to the interview stage and things don’t work out, we’ll do our best to share useful thoughts within a few days." + }, + { + "span_id": "span-1c0af5170e1b4750201dfe4a", + "text": "We genuinely wish we could give feedback to everyone who applies, but the volume makes that impossible." + }, + { + "span_id": "span-632199e6fd1583933f430f56", + "text": "We’d rather be honest about that than promise something we can’t deliver." + }, + { + "span_id": "span-ad84b987981e570873dc85a1", + "text": "Can I reapply if I was unsuccessful the first time? expand_more Yes, though we’d gently suggest waiting at least a year before reapplying for the same role." + }, + { + "span_id": "span-78d889ae4e767e3de5e7d4f4", + "text": "If we come across a different opening that feels like a better fit for your background, we’ll reach out." + } + ] + }, + { + "chunk_id": "tai-65958e1371b7d8e6c4734960", + "text": "Frequently asked questions.\nthe best way to catch new roles as they open. Does Towards AI pay for travel costs for on-site interviews? expand_more Most of our interviews happen over Google Meet or Zoom, so travel usually isn’t needed. On the rare occasion we invite you to meet in person, we’re happy to cover reasonable travel and accommodation. Just let us know what you need. I’m a recruiter. Can I help Towards AI find more talent? expand_more Thanks for the offer, and we mean that genuinely. For now we’ve built our own trusted networks for hiring, so we rarely need external support. If that changes, we work with a small list of preferred partners that we review regularly. How to apply\n\nReady to build?\nSend your LinkedIn, CV, and a short note on why this role interests you. We read every email and reply within a week. Email denis@towardsai.net arrow_forward Competitive comp + equity.", + "heading": "Frequently asked questions.", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-041c2f652461830fc08a9c40", + "text": "Does Towards AI pay for travel costs for on-site interviews? expand_more Most of our interviews happen over Google Meet or Zoom, so travel usually isn’t needed." + }, + { + "span_id": "span-3f0dd98d933ba2afe155cd98", + "text": "On the rare occasion we invite you to meet in person, we’re happy to cover reasonable travel and accommodation." + }, + { + "span_id": "span-ce91c856e5516a40e3e2ea72", + "text": "Just let us know what you need." + }, + { + "span_id": "span-240b4306f974ac3215848101", + "text": "I’m a recruiter." + }, + { + "span_id": "span-ab4f3070ef6f3979f09d6352", + "text": "Can I help Towards AI find more talent? expand_more Thanks for the offer, and we mean that genuinely." + }, + { + "span_id": "span-057c4184901ddf011f139e2f", + "text": "For now we’ve built our own trusted networks for hiring, so we rarely need external support." + }, + { + "span_id": "span-947e02167158464145be6ce4", + "text": "If that changes, we work with a small list of preferred partners that we review regularly." + }, + { + "span_id": "span-47aba23ba24e720eb0f1fe43", + "text": "How to apply" + }, + { + "span_id": "span-c0e7d540d8dd6bc707bc5d68", + "text": "Send your LinkedIn, CV, and a short note on why this role interests you." + }, + { + "span_id": "span-e8492aef010ebfa0edfac895", + "text": "We read every email and reply within a week." + }, + { + "span_id": "span-a65a7a9b0bb49f3661bb53aa", + "text": "Email denis@towardsai.net arrow_forward Competitive comp + equity." + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "55811a963776325c8077c9e83b7470385cdd292a9959cd0fcbdcde52fd5d65aa", + "content_hash": "55811a963776325c8077c9e83b7470385cdd292a9959cd0fcbdcde52fd5d65aa", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/valuecreation/careers/" + ], + "evidence_hash": "7f4e5045561824e9b79fa72d34eb9e394c17533f5fa991d1295efa3448ad2714" + }, + { + "discovered_url": "https://towardsai.com/valuecreation/deployment-strategist/", + "url": "https://towardsai.com/valuecreation/deployment-strategist/", + "canonical_url": "https://towardsai.com/valuecreation/deployment-strategist/", + "host": "towardsai.com", + "path": "/valuecreation/deployment-strategist", + "kind": "career", + "offer_id": null, + "entity_id": "page:towardsai.com/valuecreation/deployment-strategist", + "title": "Deployment Strategist · Towards AI Careers", + "review_title": "", + "reviewed_summary": "", + "meta_description": "Join Towards AI as a Deployment Strategist, turning business problems in PE-backed companies into shipped AI workflows, pilots, and production systems.", + "headings": [ + "Deployment Strategist", + "Tell us what you’ve shipped ." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Deployment Strategist\n\nDeployment Strategist\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Co-founders 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. We hire only top-tier talent. You will own client engagements at PE-backed portfolio companies (portcos) end-to-end, from first scoping conversation through to a live system the client team runs without us. You map the workflow, identify the highest-leverage automations, scope the solution, run delivery, and manage stakeholders across PE partners, portco C-level, and technical teams. You sit between investors, portco operating teams, and our AI engineers. You bring deep workflow understanding, commercial judgment, and operator instincts; the engineers bring technical depth. Together you design AI solutions that actually get implemented and create measurable value. Track record so far includes engagements with Nviya-Prime, Maoki, Europol, NYPL, J.P. Morgan, and Intel x Activeloop, and a growing book of PE and portfolio company work. 02 A recent example Nviya-Prime We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech founded by a 30-year ex-Accenture / Booz Allen / EY partner. Three coordinated sub-agent services orchestrate ~800 prompts per client analysis to replicate institutional-grade banker reasoning, with 24/7 predictive scanning that surfaces materiality-scored next-best-actions for live market events. Currently shipping into live demos with global banks across Europe. The strategist on this engagement owns the founder relationship, frames the value toward banking clients, and works alongside our engineers on the system design that connects technical capability to commercial outcome. 03 Where you will work Client-facing and senior. Significant time with PE clients, portfolio companies, and finance firms, with structured residencies on-site for kickoff, key delivery moments, and major reviews. Other work runs from our team. 04 What you'll do - Own deployment engagements at PE-backed portcos end-to-end: scoping, delivery, rollout, handover, outcome. - Map workflows, identify high-impact automations, and scope solutions calibrated to portco constraints and PE hold-period economics. - Co-design AI solutions with our senior AI engineers and forward deployed engineers (FDEs). You bring deep PE and finance domain expertise; they bring technical depth. Together you shape the workflow target, the architecture trade-offs, the success metrics, and the rollout plan. - Run delivery from kickoff to live system. Hold the engineering pod accountable to timeline, scope, quality, and adoption. - Manage stakeholders across PE partners, portco C-level executives, and technical teams. - Build sequenced AI value-creation plans for portfolio companies. Quick wins and longer plays, calibrated to hold period and earnings before interest, taxes, depreciation, and amortization (EBITDA) contribution. - Run cross-portfolio programs. Spot reusable patterns, template winning solutions, roll them across multiple portfolio companies. - Lead client education for non-technical executives. Demystify model capability, sequence priorities, calibrate expectations. - Help originate new client work across PE firms, family offices, investment platforms, and finance firms. - Help build the firm's PE practice intellectual property (IP): diligence frameworks, value-creation playbooks, sector points of view, executive workshops, and reusable AI assets. 05 AI-native working style This role requires daily, fluent use of frontier AI tools across the bulk of your professional output. You should already be a heavy user of Claude, ChatGPT, and similar systems for research, drafting, analysis, memo preparation, deck building, market scanning, due diligence, and synthesis. You know which model to reach for and when, how to structure context and prompts, how to read outputs critically, and how to combine AI speed with your own commercial judgment. AI-native strategists outpace traditional consultants and operating partners on most knowledge work, and that gap is the bar for this role. You will also design reusable AI assets that compound your effectiveness and your clients' over time: custom Claude Skills, agent skills, prompt libraries, market-mapping workflows, diligence templates, scoping playbooks, and executive briefing patterns. You bring deep domain expertise into structured AI assets the firm and clients can reuse. Where useful, you work alongside our engineers to turn your patterns into agents, sub-agents, or Model Context Protocol (MCP) servers running in production. A meaningful share of the value you deliver to PE and portfolio company clients will be teaching them this same way of working: how to use frontier AI tools safely, how to structure workflows, how to review outputs critically, how to avoid false confidence, and how to combine AI leverage with human expertise. Many of our clients need this education as urgently as they need new engineering. 06 Who you are - 2+ years at a tier-1 institution: top-tier strategy consulting (McKinsey, BCG, Bain), private equity, investment banking, value creation, portfolio operations, or PE-adjacent advisory. - Consistently performed at a high level throughout your career so far. - You take pride in measurable impact at real companies, not in slide decks. - Comfortable owning client relationships and engagement outcomes. - Real fluency with AI capability today. You do not need to write production code; you should be able to read an architecture, push back on an engineering proposal, co-design a solution with engineers, and call out an over-engineered or under-scoped plan. - Strong commercial instincts. You can scope an engagement, defend a price, and structure a multi-quarter program. - Operator judgment. You have helped management teams change how work gets done, not only delivered slides. - Strong written and verbal communication. You can write a board-ready memo and run a workshop with non-technical executives. - Daily, expert AI user. You work with Claude, ChatGPT, and similar tools across most of your professional output, design reusable prompts and skills, and can teach others to do the same. 07 Useful but not required - Direct experience inside a PE firm, portfolio operations team, or value-creation function. - Sector depth in financial services, asset management, insurance, or B2B services. - An existing book of senior relationships across PE, investment platforms, or finance firms. - MBA, CFA, or comparable analytical signal. - Authorship, speaking, or thought leadership on AI in finance, PE, or operations. - Hands-on experience designing Claude Skills, agent skills, MCP servers, or custom AI workflows. 08 What you'll learn - How to design and ship PE playbooks that actually get implemented inside real businesses. - How to work effectively alongside top-tier PE partners and operating partners at leading funds. - Daily fluency with the latest AI tools at the frontier of the field. - Proven AI-enabled workflows across procurement, sales, finance, document automation, and broader portco operations. - How to translate engineering trade-offs into board-ready commercial language without losing the substance. 09 Compensation Calibrated to candidate experience and seniority. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one. Apply\n\nTell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation/" + }, + { + "text": "Case studies", + "url": "https://towardsai.com/" + }, + { + "text": "Community arrow_forward", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Explore courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Talk to us", + "url": "https://towardsai.com/" + }, + { + "text": "Back to Towards AI Back to Towards AI", + "url": "https://towardsai.com/" + } + ], + "chunks": [ + { + "chunk_id": "tai-7dca38a70e00a6ecb0457e4a", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Deployment Strategist", + "heading": "Overview", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-685fc44b6e3486c28c0865a0", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-bee71ce46db5e54a094dee07", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-c56bb57dabff3a58790a5b2b", + "text": "Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-a6fb824375b83c1201e50a05", + "text": "Explore courses Talk to us" + }, + { + "span_id": "span-07d0ff592eb5ebbbf29908cc", + "text": "Towards AI / Careers - we’re hiring / Deployment Strategist" + } + ] + }, + { + "chunk_id": "tai-608d691c5e01b4412b5ea5c0", + "text": "Deployment Strategist\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Co-founders 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. We hire only top-tier talent. You will own client engagements at PE-backed portfolio companies (portcos) end-to-end, from first scoping conversation through to a live system the client team runs without us. You map the workflow, identify the highest-leverage automations, scope the solution, run delivery, and manage stakeholders across PE partners, portco C-level, and technical teams. You sit between investors, portco operating teams, and our AI engineers. You bring deep workflow understanding, commercial judgment, and operator instincts; the engineers bring technical depth. Together you design AI solutions that actually get implemented and create measurable value. Track record so far includes engagements with Nviya-Prime, Maoki, Europol, NYPL, J.P. Morgan, and Intel x Activeloop, and a growing book of PE and portfolio company work. 02 A recent example Nviya-Prime We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech founded by a 30-year ex-Accenture / Booz Allen / EY partner. Three coordinated sub-agent services orchestrate ~800 prompts per client analysis to replicate institutional-grade banker reasoning, with 24/7 predictive scanning that surfaces materiality-scored next-best-actions for live market events. Currently shipping into live demos with global banks", + "heading": "Deployment Strategist", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-043f930a4884571a038c9236", + "text": "We bring AI inside organisations." + }, + { + "span_id": "span-ff3b2fd2d0b91a922a329109", + "text": "We build the systems." + }, + { + "span_id": "span-4c7361b3499cbd3ece2311e0", + "text": "We train the people." + }, + { + "span_id": "span-e603b22b3219a1fdc90ae0af", + "text": "Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI" + }, + { + "span_id": "span-1bff2dfd76400b286c408b54", + "text": "AI Engineering / Value Creation" + }, + { + "span_id": "span-9e8254f94d3bdbb590a97718", + "text": "London or Hybrid" + }, + { + "span_id": "span-7cbf6a91feb66601e54c7562", + "text": "Apply by" + }, + { + "span_id": "span-9f1a1e6445b64696e2e28268", + "text": "Reports to" + }, + { + "span_id": "span-f7fe0f734c261a1317838ef1", + "text": "Co-founders" + }, + { + "span_id": "span-0c591d1c0819c1630718d4a9", + "text": "01 The opportunity" + }, + { + "span_id": "span-4f0ddab24f309127ccef0555", + "text": "We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong." + }, + { + "span_id": "span-6e31960fa536ffffee998f6b", + "text": "We hire only top-tier talent." + }, + { + "span_id": "span-fcf3d1fcf4c3f2d2f8ffa8c1", + "text": "You will own client engagements at PE-backed portfolio companies (portcos) end-to-end, from first scoping conversation through to a live system the client team runs without us." + }, + { + "span_id": "span-2c68968e0a7797dd2839a0e1", + "text": "You map the workflow, identify the highest-leverage automations, scope the solution, run delivery, and manage stakeholders across PE partners, portco C-level, and technical teams." + }, + { + "span_id": "span-3d43c83cc6665dbd088dfbd2", + "text": "You sit between investors, portco operating teams, and our AI engineers." + }, + { + "span_id": "span-69bb40b13b79034a79775f53", + "text": "You bring deep workflow understanding, commercial judgment, and operator instincts; the engineers bring technical depth." + }, + { + "span_id": "span-2c9faef37dd45fcda310c7c9", + "text": "Together you design AI solutions that actually get implemented and create measurable value." + }, + { + "span_id": "span-c79c6727de7b43cd0b725140", + "text": "Track record so far includes engagements with Nviya-Prime, Maoki, Europol, NYPL, J.P." + }, + { + "span_id": "span-fd04ee8c9cb92a6b0decc2bc", + "text": "Morgan, and Intel x Activeloop, and a growing book of PE and portfolio company work." + }, + { + "span_id": "span-2ca7d35d7134153759579b39", + "text": "02 A recent example" + }, + { + "span_id": "span-08cbb106ee7f512ea59506b2", + "text": "Nviya-Prime" + }, + { + "span_id": "span-2c2a3b2e6de6d33aeb64b957", + "text": "We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech founded by a 30-year ex-Accenture / Booz Allen / EY partner." + }, + { + "span_id": "span-a673595abb8e5b7411525602", + "text": "Three coordinated sub-agent services orchestrate ~800 prompts per client analysis to replicate institutional-grade banker reasoning, with 24/7 predictive scanning that surfaces materiality-scored next-best-actions for live market events." + } + ] + }, + { + "chunk_id": "tai-7be0172a936f700493052d2e", + "text": "Deployment Strategist\nacross Europe. The strategist on this engagement owns the founder relationship, frames the value toward banking clients, and works alongside our engineers on the system design that connects technical capability to commercial outcome. 03 Where you will work Client-facing and senior. Significant time with PE clients, portfolio companies, and finance firms, with structured residencies on-site for kickoff, key delivery moments, and major reviews. Other work runs from our team. 04 What you'll do - Own deployment engagements at PE-backed portcos end-to-end: scoping, delivery, rollout, handover, outcome. - Map workflows, identify high-impact automations, and scope solutions calibrated to portco constraints and PE hold-period economics. - Co-design AI solutions with our senior AI engineers and forward deployed engineers (FDEs). You bring deep PE and finance domain expertise; they bring technical depth. Together you shape the workflow target, the architecture trade-offs, the success metrics, and the rollout plan. - Run delivery from kickoff to live system. Hold the engineering pod accountable to timeline, scope, quality, and adoption. - Manage stakeholders across PE partners, portco C-level executives, and technical teams. - Build sequenced AI value-creation plans for portfolio companies. Quick wins and longer plays, calibrated to hold period and earnings before interest, taxes, depreciation, and amortization (EBITDA) contribution. - Run cross-portfolio programs. Spot reusable patterns, template winning solutions, roll them across multiple portfolio companies. - Lead client education for non-technical executives. Demystify model capability, sequence priorities, calibrate expectations. - Help originate new client work across PE firms, family offices,", + "heading": "Deployment Strategist", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-bb46c8d803aafefa75b55b2d", + "text": "The strategist on this engagement owns the founder relationship, frames the value toward banking clients, and works alongside our engineers on the system design that connects technical capability to commercial outcome." + }, + { + "span_id": "span-2b6d6a2d2809e2fcdd5be76a", + "text": "03 Where you will work" + }, + { + "span_id": "span-2b1333316b404084e772b4d0", + "text": "Client-facing and senior." + }, + { + "span_id": "span-a17b2002f55c5347b2c42234", + "text": "Significant time with PE clients, portfolio companies, and finance firms, with structured residencies on-site for kickoff, key delivery moments, and major reviews." + }, + { + "span_id": "span-232b13cf68d56911f4689e99", + "text": "Other work runs from our team." + }, + { + "span_id": "span-4cc3abbf5662795fae0ee8fe", + "text": "04 What you'll do" + }, + { + "span_id": "span-e62cb4126a68c81d358873bd", + "text": "- Own deployment engagements at PE-backed portcos end-to-end: scoping, delivery, rollout, handover, outcome." + }, + { + "span_id": "span-6b5d7c4f5bf4290ee436f5c3", + "text": "- Map workflows, identify high-impact automations, and scope solutions calibrated to portco constraints and PE hold-period economics." + }, + { + "span_id": "span-7d7aee297132a3800b63d910", + "text": "- Co-design AI solutions with our senior AI engineers and forward deployed engineers (FDEs)." + }, + { + "span_id": "span-b9fd8a37dba91b92013131ca", + "text": "You bring deep PE and finance domain expertise; they bring technical depth." + }, + { + "span_id": "span-9d56701d8a1881401c4e8417", + "text": "Together you shape the workflow target, the architecture trade-offs, the success metrics, and the rollout plan." + }, + { + "span_id": "span-3d04be1a2730852fde0b47c7", + "text": "- Run delivery from kickoff to live system." + }, + { + "span_id": "span-073c2409892811895b06bf10", + "text": "Hold the engineering pod accountable to timeline, scope, quality, and adoption." + }, + { + "span_id": "span-204f1272274263d44310b8dc", + "text": "- Manage stakeholders across PE partners, portco C-level executives, and technical teams." + }, + { + "span_id": "span-40b91f913a14531d2b723a84", + "text": "- Build sequenced AI value-creation plans for portfolio companies." + }, + { + "span_id": "span-e1ef6cbfca03e8144f39da89", + "text": "Quick wins and longer plays, calibrated to hold period and earnings before interest, taxes, depreciation, and amortization (EBITDA) contribution." + }, + { + "span_id": "span-a2a6db0bd3e697eaac00ac5f", + "text": "- Run cross-portfolio programs." + }, + { + "span_id": "span-285b14f07f2f66d5b0407f63", + "text": "Spot reusable patterns, template winning solutions, roll them across multiple portfolio companies." + }, + { + "span_id": "span-64fcde128694a357b282a672", + "text": "- Lead client education for non-technical executives." + }, + { + "span_id": "span-335a7857c81c53ed4a137ca1", + "text": "Demystify model capability, sequence priorities, calibrate expectations." + } + ] + }, + { + "chunk_id": "tai-da7159ad233aa28efa496074", + "text": "Deployment Strategist\ninvestment platforms, and finance firms. - Help build the firm's PE practice intellectual property (IP): diligence frameworks, value-creation playbooks, sector points of view, executive workshops, and reusable AI assets. 05 AI-native working style This role requires daily, fluent use of frontier AI tools across the bulk of your professional output. You should already be a heavy user of Claude, ChatGPT, and similar systems for research, drafting, analysis, memo preparation, deck building, market scanning, due diligence, and synthesis. You know which model to reach for and when, how to structure context and prompts, how to read outputs critically, and how to combine AI speed with your own commercial judgment. AI-native strategists outpace traditional consultants and operating partners on most knowledge work, and that gap is the bar for this role. You will also design reusable AI assets that compound your effectiveness and your clients' over time: custom Claude Skills, agent skills, prompt libraries, market-mapping workflows, diligence templates, scoping playbooks, and executive briefing patterns. You bring deep domain expertise into structured AI assets the firm and clients can reuse. Where useful, you work alongside our engineers to turn your patterns into agents, sub-agents, or Model Context Protocol (MCP) servers running in production. A meaningful share of the value you deliver to PE and portfolio company clients will be teaching them this same way of working: how to use frontier AI tools safely, how to structure workflows, how to review outputs critically, how to avoid false confidence, and how to combine AI leverage with human expertise. Many of our clients need this education as urgently as they need new engineering. 06 Who you are - 2+ years", + "heading": "Deployment Strategist", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-c7297dbd45742613d98b4ffc", + "text": "- Help build the firm's PE practice intellectual property (IP): diligence frameworks, value-creation playbooks, sector points of view, executive workshops, and reusable AI assets." + }, + { + "span_id": "span-0e99912153d8c610aad722d4", + "text": "05 AI-native working style" + }, + { + "span_id": "span-08c6debfc7d8a3f3e1845a84", + "text": "This role requires daily, fluent use of frontier AI tools across the bulk of your professional output." + }, + { + "span_id": "span-1db034c2bd1ed06068e463c0", + "text": "You should already be a heavy user of Claude, ChatGPT, and similar systems for research, drafting, analysis, memo preparation, deck building, market scanning, due diligence, and synthesis." + }, + { + "span_id": "span-31b3053620204926aac16fe9", + "text": "You know which model to reach for and when, how to structure context and prompts, how to read outputs critically, and how to combine AI speed with your own commercial judgment." + }, + { + "span_id": "span-604352c947970979cd3809b9", + "text": "AI-native strategists outpace traditional consultants and operating partners on most knowledge work, and that gap is the bar for this role." + }, + { + "span_id": "span-880699efb56ab7278f1ca6ae", + "text": "You will also design reusable AI assets that compound your effectiveness and your clients' over time: custom Claude Skills, agent skills, prompt libraries, market-mapping workflows, diligence templates, scoping playbooks, and executive briefing patterns." + }, + { + "span_id": "span-00572f90567f90698b183aa6", + "text": "You bring deep domain expertise into structured AI assets the firm and clients can reuse." + }, + { + "span_id": "span-bcbe79405052a8c845cd8ffa", + "text": "Where useful, you work alongside our engineers to turn your patterns into agents, sub-agents, or Model Context Protocol (MCP) servers running in production." + }, + { + "span_id": "span-731030338c096acd4db8424c", + "text": "A meaningful share of the value you deliver to PE and portfolio company clients will be teaching them this same way of working: how to use frontier AI tools safely, how to structure workflows, how to review outputs critically, how to avoid false confidence, and how to combine AI leverage with human expertise." + }, + { + "span_id": "span-f0481cbca427df71c2ad3bd3", + "text": "Many of our clients need this education as urgently as they need new engineering." + }, + { + "span_id": "span-e3da74ea4c1adf0ada1c6b7b", + "text": "06 Who you are" + } + ] + }, + { + "chunk_id": "tai-4e1c36dc8caf299a683e78a2", + "text": "Deployment Strategist\nat a tier-1 institution: top-tier strategy consulting (McKinsey, BCG, Bain), private equity, investment banking, value creation, portfolio operations, or PE-adjacent advisory. - Consistently performed at a high level throughout your career so far. - You take pride in measurable impact at real companies, not in slide decks. - Comfortable owning client relationships and engagement outcomes. - Real fluency with AI capability today. You do not need to write production code; you should be able to read an architecture, push back on an engineering proposal, co-design a solution with engineers, and call out an over-engineered or under-scoped plan. - Strong commercial instincts. You can scope an engagement, defend a price, and structure a multi-quarter program. - Operator judgment. You have helped management teams change how work gets done, not only delivered slides. - Strong written and verbal communication. You can write a board-ready memo and run a workshop with non-technical executives. - Daily, expert AI user. You work with Claude, ChatGPT, and similar tools across most of your professional output, design reusable prompts and skills, and can teach others to do the same. 07 Useful but not required - Direct experience inside a PE firm, portfolio operations team, or value-creation function. - Sector depth in financial services, asset management, insurance, or B2B services. - An existing book of senior relationships across PE, investment platforms, or finance firms. - MBA, CFA, or comparable analytical signal. - Authorship, speaking, or thought leadership on AI in finance, PE, or operations. - Hands-on experience designing Claude Skills, agent skills, MCP servers, or custom AI workflows. 08 What you'll learn - How to design and ship PE playbooks that", + "heading": "Deployment Strategist", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-a5e36254b1deb0758f314519", + "text": "- Consistently performed at a high level throughout your career so far." + }, + { + "span_id": "span-6818fd3c5057e2073070ca8f", + "text": "- You take pride in measurable impact at real companies, not in slide decks." + }, + { + "span_id": "span-5f0e86257c4cb291837e66c3", + "text": "- Comfortable owning client relationships and engagement outcomes." + }, + { + "span_id": "span-34816cd621ce9a7b9541c123", + "text": "- Real fluency with AI capability today." + }, + { + "span_id": "span-e3e5eaa3ea9d03b8f35ddb76", + "text": "You do not need to write production code; you should be able to read an architecture, push back on an engineering proposal, co-design a solution with engineers, and call out an over-engineered or under-scoped plan." + }, + { + "span_id": "span-6b8caef5dab702a0e50ac64d", + "text": "- Strong commercial instincts." + }, + { + "span_id": "span-6e4235925870531b2152c033", + "text": "You can scope an engagement, defend a price, and structure a multi-quarter program." + }, + { + "span_id": "span-68e9bb19910703e66210ba9c", + "text": "- Operator judgment." + }, + { + "span_id": "span-d2568cbf3ff63cddd46a63c7", + "text": "You have helped management teams change how work gets done, not only delivered slides." + }, + { + "span_id": "span-bddfb345c987d7b997d1778d", + "text": "- Strong written and verbal communication." + }, + { + "span_id": "span-73f6515adee183d0cabe0a64", + "text": "You can write a board-ready memo and run a workshop with non-technical executives." + }, + { + "span_id": "span-864f288c1ff2d1fb6df663bd", + "text": "- Daily, expert AI user." + }, + { + "span_id": "span-8283d84ee81e1fb667f52356", + "text": "You work with Claude, ChatGPT, and similar tools across most of your professional output, design reusable prompts and skills, and can teach others to do the same." + }, + { + "span_id": "span-c89789e4dad2811843098f16", + "text": "07 Useful but not required" + }, + { + "span_id": "span-bf71568a03cf8ab7247d8880", + "text": "- Direct experience inside a PE firm, portfolio operations team, or value-creation function." + }, + { + "span_id": "span-ab3209a1fc081f24410f7a9e", + "text": "- Sector depth in financial services, asset management, insurance, or B2B services." + }, + { + "span_id": "span-5e07fe35be0de9b5249557d1", + "text": "- An existing book of senior relationships across PE, investment platforms, or finance firms." + }, + { + "span_id": "span-889b1be014c7388e3a3ceb06", + "text": "- MBA, CFA, or comparable analytical signal." + }, + { + "span_id": "span-28677716bfbf440ec7cffd8f", + "text": "- Authorship, speaking, or thought leadership on AI in finance, PE, or operations." + }, + { + "span_id": "span-424e1215afaa94d4a79e7c74", + "text": "- Hands-on experience designing Claude Skills, agent skills, MCP servers, or custom AI workflows." + }, + { + "span_id": "span-a017b35be3e81e6c6a6a4ff9", + "text": "08 What you'll learn" + } + ] + }, + { + "chunk_id": "tai-5e7a24fd60918c567e52bf57", + "text": "Deployment Strategist\nactually get implemented inside real businesses. - How to work effectively alongside top-tier PE partners and operating partners at leading funds. - Daily fluency with the latest AI tools at the frontier of the field. - Proven AI-enabled workflows across procurement, sales, finance, document automation, and broader portco operations. - How to translate engineering trade-offs into board-ready commercial language without losing the substance. 09 Compensation Calibrated to candidate experience and seniority. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one. Apply", + "heading": "Deployment Strategist", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-a6c5ff2af3269fbfd68b071b", + "text": "- How to work effectively alongside top-tier PE partners and operating partners at leading funds." + }, + { + "span_id": "span-254b973b4fdd867aae3f6707", + "text": "- Daily fluency with the latest AI tools at the frontier of the field." + }, + { + "span_id": "span-878ec371910e49661853539c", + "text": "- Proven AI-enabled workflows across procurement, sales, finance, document automation, and broader portco operations." + }, + { + "span_id": "span-21052e6f2e2230f5ab9ddadb", + "text": "- How to translate engineering trade-offs into board-ready commercial language without losing the substance." + }, + { + "span_id": "span-1c8f66082c88e53dbb1ba1cb", + "text": "09 Compensation" + }, + { + "span_id": "span-9b0cbb6d1f2b1ad5242393fc", + "text": "Calibrated to candidate experience and seniority." + }, + { + "span_id": "span-1e78dceed407c415930f5850", + "text": "About Towards AI" + }, + { + "span_id": "span-132016d22f0ba6fc46e6034f", + "text": "Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production." + }, + { + "span_id": "span-8aa3d71f52e9c41a516cca80", + "text": "Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries." + }, + { + "span_id": "span-d94980bc8ea66936a8f24938", + "text": "Co-founded by Louie Peters (ex-J.P." + }, + { + "span_id": "span-da0258872467118994f5c23f", + "text": "Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal)." + }, + { + "span_id": "span-de20a4b4f99079febdb55bea", + "text": "Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field." + }, + { + "span_id": "span-bfa099868aa986c70a9ba4b1", + "text": "In a pod you own a real system end to end, in production, from day one." + } + ] + }, + { + "chunk_id": "tai-335d61643af14e3457564033", + "text": "Tell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "heading": "Tell us what you’ve shipped .", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-301081c90568ce2d4656d0d9", + "text": "Concrete write-ups beat resumes." + }, + { + "span_id": "span-40a7eaa514b7652097ee9377", + "text": "We read every application." + }, + { + "span_id": "span-e746bc00035aeee9e6baf84f", + "text": "If we want to talk, you’ll hear from us inside a week." + }, + { + "span_id": "span-5ce07e5523b8834c5d021189", + "text": "Full name" + }, + { + "span_id": "span-09a4e5588d17ca4ad9f2bc01", + "text": "Country you’re based in" + }, + { + "span_id": "span-2b62435c4e2a4a651c7c7a90", + "text": "Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role" + }, + { + "span_id": "span-80fb62bdebdc477aac5b00aa", + "text": "LinkedIn profile URL" + }, + { + "span_id": "span-fccb220848c6fc572af143d0", + "text": "CV / Resume (PDF preferred)" + }, + { + "span_id": "span-ef15d608e8ff52ee9b6c5bbb", + "text": "Link to your best public work" + }, + { + "span_id": "span-953a45f9a083a33b1b791d2b", + "text": "Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines)" + }, + { + "span_id": "span-4be4e5bdac7ba2824b7c0f4c", + "text": "How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work?" + }, + { + "span_id": "span-579c08f9ccf283c086e4536e", + "text": "Submit application arrow_forward" + }, + { + "span_id": "span-122c60f1ce94ad59963c94cf", + "text": "Apply for this role Apply for this role arrow_forward" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "0d03cc4a8735d33eab689513bab3e754e1dc1930edb99f3f40c7fd5cb7d296ed", + "content_hash": "0d03cc4a8735d33eab689513bab3e754e1dc1930edb99f3f40c7fd5cb7d296ed", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/valuecreation/deployment-strategist/" + ], + "evidence_hash": "367683b4998fd40958594017fed02b6970393272e152ae7bfdabdfe3717db5ad" + }, + { + "discovered_url": "https://towardsai.com/valuecreation/junior-ai-engineer/", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/", + "canonical_url": "https://towardsai.com/valuecreation/junior-ai-engineer/", + "host": "towardsai.com", + "path": "/valuecreation/junior-ai-engineer", + "kind": "career", + "offer_id": null, + "entity_id": "page:towardsai.com/valuecreation/junior-ai-engineer", + "title": "Junior AI Engineer · Towards AI Careers", + "review_title": "", + "reviewed_summary": "", + "meta_description": "Join Towards AI as a Junior AI Engineer, shipping production AI features with senior engineers across retrieval, agents, automation, and evaluation.", + "headings": [ + "Junior AI Engineer", + "Tell us what you’ve shipped ." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Junior AI Engineer\n\nJunior AI Engineer\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Senior AI Engineer 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. We hire only top-tier talent. Our 15-person engineering team ships agentic systems, retrieval-augmented generation (RAG) pipelines, custom copilots, and document automation into clients including Nviya-Prime, Maoki, Europol, NYPL, J.P. Morgan, and Intel x Activeloop. This role is for an early-career engineer who wants to grow inside a serious AI engineering team. You will work alongside our senior engineers on real production LLM systems and learn the full engineering loop. The role is fully internal by design: no client calls, no executive conversations, no on-site work. Senior engineers and the deployment strategist own all of that. Your job is to build, evaluate, and ship. Over time, as you grow into a senior engineer, there is a clear path into more client-facing or forward-deployed work. 02 A recent project example Maoki Our team built an automated commercial mortgage document processing pipeline for Maoki, a Canadian commercial mortgage brokerage. The system handles 50+ document types in English and French, runs three independent extraction pipelines with human-in-the-loop review, integrates conditionally across seven Salesforce objects with never-overwrite semantics, and runs Canadian-resident inference on Vertex AI. A junior engineer on a project like this would contribute to retrieval components, eval harnesses, integration code, and pipeline reliability work, with senior engineering oversight on architecture and all client interaction. 03 What you will do - Build and maintain components inside production LLM applications: retrieval pipelines, agent workflows, eval harnesses, integrations, internal tools. - Own well-scoped tickets end-to-end: design, implementation, tests, evals, code review, deployment. - Implement and improve RAG pipelines: parsing, chunking, indexing, retrieval, reranking, grounding, source attribution. - Build and maintain agent components: tool calling, structured outputs, orchestration steps, retries, human review hooks. - Build evaluation suites: golden datasets, regression tests, retrieval and generation metrics, observability hooks. - Integrate AI systems with client data sources, document repositories, customer relationship management (CRM) systems, knowledge bases, internal application programming interfaces (APIs). - Write clean, tested, documented Python; contribute to code review and engineering discussion. - Contribute to reusable firm assets: prompt libraries, agent skills, eval harnesses, reference patterns, internal documentation. 04 AI-native working style You should already be a daily, native user of agentic coding tools: Claude Code, Codex, Cursor, or similar. Not occasional experiments. We expect them in your default workflow for planning, codebase exploration, implementation, refactoring, testing, debugging, and documentation. We hire on this. The standard is intelligent supervision, not blind delegation. You give the agent the right context, constrain the task, inspect what it produces, run the tests, and catch weak assumptions before committing. You never ship code you have not understood. Beyond coding tools, you will help build reusable AI infrastructure the team uses across engagements: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, workflow templates. 05 Required qualifications - 1-3 years professional software engineering experience, or equivalent strong evidence (open-source work, published projects, hackathons, technical writing). - At least one shipped LLM project beyond simple chat: a working RAG pipeline, an agent with real tool use, an internal copilot, a document analysis system, or comparable. - Strong Python. Comfort with APIs, JSON, async, packaging, testing. - Working knowledge of OpenAI, Anthropic, Gemini, or comparable LLM APIs. - Practical understanding of RAG: embeddings, chunking, vector search, reranking, basic retrieval evaluation. - Comfort with Git, Docker, basic cloud deployment, Linux command line. - Habit of evaluating LLM outputs rather than trusting them: written test cases, golden sets, regression checks. - Daily use of agentic coding tools (Claude Code, Codex, Cursor) for real work, with careful inspection of what they produce. - Strong written communication. Clear pull request (PR) descriptions, clear questions, clear documentation. 06 Useful but not required - Familiarity with LangChain, LangGraph, LlamaIndex, LangSmith, Langfuse, Braintrust, MCP servers, or OpenAI Agents SDK. - Vector databases such as Pinecone, Weaviate, Qdrant, Chroma, or pgvector. - TypeScript, React, Next.js, Postgres, or Supabase. - Domain interest in finance, investment, operations, sales, or customer support. - Experience with fine-tuning, distillation, or open-weight models. 07 What you will learn - How to design, build, evaluate, and ship production LLM systems, not demos. - How to debug agent failures, fix retrieval quality issues, and harden systems for real use. - How to evaluate models, prompts, and pipelines with discipline rather than vibes. - How to use agentic coding tools at a senior level: planning, supervision, intelligent rejection. - How AI engineering translates into real client value, observed through the senior engineers and strategists you support. 08 Who we are looking for An early-career engineer with strong technical fundamentals, real curiosity about AI engineering as a craft, and the temperament for a high-bar environment. You care about correctness, evaluation, and useful outputs more than impressive vocabulary. You read other people's code carefully, ask sharp questions, ship fast under intelligent supervision of agentic tools, and grow fast. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one. Apply\n\nTell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation/" + }, + { + "text": "Case studies", + "url": "https://towardsai.com/" + }, + { + "text": "Community arrow_forward", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Explore courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Talk to us", + "url": "https://towardsai.com/" + }, + { + "text": "Back to Towards AI Back to Towards AI", + "url": "https://towardsai.com/" + } + ], + "chunks": [ + { + "chunk_id": "tai-2175e9fd763b47af2951d051", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Junior AI Engineer", + "heading": "Overview", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-957db5c7729209ae73ef58fb", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-67cda50741bfdd3b234d299f", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-6dbc4d385b90f572ef04a2fe", + "text": "Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-101ab37f0f3faa9b54ccd305", + "text": "Explore courses Talk to us" + }, + { + "span_id": "span-3b3de2fffc8580a32c337b1c", + "text": "Towards AI / Careers - we’re hiring / Junior AI Engineer" + } + ] + }, + { + "chunk_id": "tai-dd8300b7e568f4650c6c6e95", + "text": "Junior AI Engineer\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Senior AI Engineer 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. We hire only top-tier talent. Our 15-person engineering team ships agentic systems, retrieval-augmented generation (RAG) pipelines, custom copilots, and document automation into clients including Nviya-Prime, Maoki, Europol, NYPL, J.P. Morgan, and Intel x Activeloop. This role is for an early-career engineer who wants to grow inside a serious AI engineering team. You will work alongside our senior engineers on real production LLM systems and learn the full engineering loop. The role is fully internal by design: no client calls, no executive conversations, no on-site work. Senior engineers and the deployment strategist own all of that. Your job is to build, evaluate, and ship. Over time, as you grow into a senior engineer, there is a clear path into more client-facing or forward-deployed work. 02 A recent project example Maoki Our team built an automated commercial mortgage document processing pipeline for Maoki, a Canadian commercial mortgage brokerage. The system handles 50+ document types in English and French, runs three independent extraction pipelines with human-in-the-loop review, integrates conditionally across seven Salesforce objects with never-overwrite semantics, and runs Canadian-resident inference on Vertex AI. A junior engineer on a project like this would contribute", + "heading": "Junior AI Engineer", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-562090a909d0cabab46621aa", + "text": "We bring AI inside organisations." + }, + { + "span_id": "span-ec2d021ca12559ec2bc4349b", + "text": "We build the systems." + }, + { + "span_id": "span-1816d8f6c5ec3fb4fe7d4705", + "text": "We train the people." + }, + { + "span_id": "span-89e749cddf11cdaae906772f", + "text": "Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI" + }, + { + "span_id": "span-80d85b5046bfc99124af5007", + "text": "AI Engineering / Value Creation" + }, + { + "span_id": "span-af46f239615af468ea7a37a3", + "text": "London or Hybrid" + }, + { + "span_id": "span-93bbfcfa9d67e819ac0b1eef", + "text": "Apply by" + }, + { + "span_id": "span-fa6cebbbe9bfee037ce5d1f8", + "text": "Reports to" + }, + { + "span_id": "span-66842157ac1212f2599a0f38", + "text": "Senior AI Engineer" + }, + { + "span_id": "span-3d038e4735f984a59c7a2883", + "text": "01 The opportunity" + }, + { + "span_id": "span-67ccf78e34b35f026f2dc8c0", + "text": "We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong." + }, + { + "span_id": "span-143dec8d7a64575e38a2c96b", + "text": "We hire only top-tier talent." + }, + { + "span_id": "span-9d0d312241b1ddb94f77cbed", + "text": "Our 15-person engineering team ships agentic systems, retrieval-augmented generation (RAG) pipelines, custom copilots, and document automation into clients including Nviya-Prime, Maoki, Europol, NYPL, J.P." + }, + { + "span_id": "span-8d9718ce2d70540aae12e282", + "text": "Morgan, and Intel x Activeloop." + }, + { + "span_id": "span-3ce51bc12755cf61568fe086", + "text": "This role is for an early-career engineer who wants to grow inside a serious AI engineering team." + }, + { + "span_id": "span-3834c17cef51bdd8bec05959", + "text": "You will work alongside our senior engineers on real production LLM systems and learn the full engineering loop." + }, + { + "span_id": "span-a11ad4279857497600162755", + "text": "The role is fully internal by design: no client calls, no executive conversations, no on-site work." + }, + { + "span_id": "span-37d18c493b958d27a2924843", + "text": "Senior engineers and the deployment strategist own all of that." + }, + { + "span_id": "span-469ecd7e8e0b6620bdb3c669", + "text": "Your job is to build, evaluate, and ship." + }, + { + "span_id": "span-97aed7a60e471ed35ba67f7a", + "text": "Over time, as you grow into a senior engineer, there is a clear path into more client-facing or forward-deployed work." + }, + { + "span_id": "span-48af9ade29ba8f63d2ce966e", + "text": "02 A recent project example" + }, + { + "span_id": "span-f5d1b7335f98ff1637ab8db4", + "text": "Our team built an automated commercial mortgage document processing pipeline for Maoki, a Canadian commercial mortgage brokerage." + }, + { + "span_id": "span-103bcebfb3d7148548ee8ad0", + "text": "The system handles 50+ document types in English and French, runs three independent extraction pipelines with human-in-the-loop review, integrates conditionally across seven Salesforce objects with never-overwrite semantics, and runs Canadian-resident inference on Vertex AI." + } + ] + }, + { + "chunk_id": "tai-ec90fb4c2d58b6d1923a492c", + "text": "Junior AI Engineer\nto retrieval components, eval harnesses, integration code, and pipeline reliability work, with senior engineering oversight on architecture and all client interaction. 03 What you will do - Build and maintain components inside production LLM applications: retrieval pipelines, agent workflows, eval harnesses, integrations, internal tools. - Own well-scoped tickets end-to-end: design, implementation, tests, evals, code review, deployment. - Implement and improve RAG pipelines: parsing, chunking, indexing, retrieval, reranking, grounding, source attribution. - Build and maintain agent components: tool calling, structured outputs, orchestration steps, retries, human review hooks. - Build evaluation suites: golden datasets, regression tests, retrieval and generation metrics, observability hooks. - Integrate AI systems with client data sources, document repositories, customer relationship management (CRM) systems, knowledge bases, internal application programming interfaces (APIs). - Write clean, tested, documented Python; contribute to code review and engineering discussion. - Contribute to reusable firm assets: prompt libraries, agent skills, eval harnesses, reference patterns, internal documentation. 04 AI-native working style You should already be a daily, native user of agentic coding tools: Claude Code, Codex, Cursor, or similar. Not occasional experiments. We expect them in your default workflow for planning, codebase exploration, implementation, refactoring, testing, debugging, and documentation. We hire on this. The standard is intelligent supervision, not blind delegation. You give the agent the right context, constrain the task, inspect what it produces, run the tests, and catch weak assumptions before committing. You never ship code you have not", + "heading": "Junior AI Engineer", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-1a6daf1878633a32853dc990", + "text": "03 What you will do" + }, + { + "span_id": "span-ec026c5cc357d00afff3ae70", + "text": "- Build and maintain components inside production LLM applications: retrieval pipelines, agent workflows, eval harnesses, integrations, internal tools." + }, + { + "span_id": "span-32d544cdedcc3f9b0b5a7810", + "text": "- Own well-scoped tickets end-to-end: design, implementation, tests, evals, code review, deployment." + }, + { + "span_id": "span-a9806e8d8ba32c5b3083dd6e", + "text": "- Implement and improve RAG pipelines: parsing, chunking, indexing, retrieval, reranking, grounding, source attribution." + }, + { + "span_id": "span-b474c56a1c317c587a3929a0", + "text": "- Build and maintain agent components: tool calling, structured outputs, orchestration steps, retries, human review hooks." + }, + { + "span_id": "span-d12285e07e0241981300d63f", + "text": "- Build evaluation suites: golden datasets, regression tests, retrieval and generation metrics, observability hooks." + }, + { + "span_id": "span-05411c13d3e7592ac930cad1", + "text": "- Integrate AI systems with client data sources, document repositories, customer relationship management (CRM) systems, knowledge bases, internal application programming interfaces (APIs)." + }, + { + "span_id": "span-e1c76e247a39fac7e9e60747", + "text": "- Write clean, tested, documented Python; contribute to code review and engineering discussion." + }, + { + "span_id": "span-9384bdfce639a61f957712f2", + "text": "- Contribute to reusable firm assets: prompt libraries, agent skills, eval harnesses, reference patterns, internal documentation." + }, + { + "span_id": "span-286a5670793f7b304567e7b4", + "text": "04 AI-native working style" + }, + { + "span_id": "span-487dc9adf33a2171235cfbd0", + "text": "You should already be a daily, native user of agentic coding tools: Claude Code, Codex, Cursor, or similar." + }, + { + "span_id": "span-e54015470a13fb3dedf9d6b6", + "text": "Not occasional experiments." + }, + { + "span_id": "span-09e93a521bb13cc095643b77", + "text": "We expect them in your default workflow for planning, codebase exploration, implementation, refactoring, testing, debugging, and documentation." + }, + { + "span_id": "span-e8d365141819d7f02ce06e71", + "text": "We hire on this." + }, + { + "span_id": "span-dec6180099585987c1ca069e", + "text": "The standard is intelligent supervision, not blind delegation." + }, + { + "span_id": "span-8f3697071db8dc1785067bdc", + "text": "You give the agent the right context, constrain the task, inspect what it produces, run the tests, and catch weak assumptions before committing." + } + ] + }, + { + "chunk_id": "tai-79ef57199bc082142ab53eea", + "text": "Junior AI Engineer\nunderstood. Beyond coding tools, you will help build reusable AI infrastructure the team uses across engagements: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, workflow templates. 05 Required qualifications - 1-3 years professional software engineering experience, or equivalent strong evidence (open-source work, published projects, hackathons, technical writing). - At least one shipped LLM project beyond simple chat: a working RAG pipeline, an agent with real tool use, an internal copilot, a document analysis system, or comparable. - Strong Python. Comfort with APIs, JSON, async, packaging, testing. - Working knowledge of OpenAI, Anthropic, Gemini, or comparable LLM APIs. - Practical understanding of RAG: embeddings, chunking, vector search, reranking, basic retrieval evaluation. - Comfort with Git, Docker, basic cloud deployment, Linux command line. - Habit of evaluating LLM outputs rather than trusting them: written test cases, golden sets, regression checks. - Daily use of agentic coding tools (Claude Code, Codex, Cursor) for real work, with careful inspection of what they produce. - Strong written communication. Clear pull request (PR) descriptions, clear questions, clear documentation. 06 Useful but not required - Familiarity with LangChain, LangGraph, LlamaIndex, LangSmith, Langfuse, Braintrust, MCP servers, or OpenAI Agents SDK. - Vector databases such as Pinecone, Weaviate, Qdrant, Chroma, or pgvector. - TypeScript, React, Next.js, Postgres, or Supabase. - Domain interest in finance, investment, operations, sales, or customer support. - Experience with fine-tuning, distillation, or open-weight models. 07 What you will learn - How to design, build, evaluate, and ship production LLM systems,", + "heading": "Junior AI Engineer", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-af8ebb13dd74eb20b7b35e05", + "text": "Beyond coding tools, you will help build reusable AI infrastructure the team uses across engagements: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, workflow templates." + }, + { + "span_id": "span-831e0b69ffc0d34fc4d96abf", + "text": "05 Required qualifications" + }, + { + "span_id": "span-190f41ebce3244f266e97361", + "text": "- 1-3 years professional software engineering experience, or equivalent strong evidence (open-source work, published projects, hackathons, technical writing)." + }, + { + "span_id": "span-2860c5d8270f49a17f7669b6", + "text": "- At least one shipped LLM project beyond simple chat: a working RAG pipeline, an agent with real tool use, an internal copilot, a document analysis system, or comparable." + }, + { + "span_id": "span-9126bd37650912394521c679", + "text": "- Strong Python." + }, + { + "span_id": "span-fe2dc2d8029b4de8fb6de415", + "text": "Comfort with APIs, JSON, async, packaging, testing." + }, + { + "span_id": "span-d4142a295362fbef2e08fd87", + "text": "- Working knowledge of OpenAI, Anthropic, Gemini, or comparable LLM APIs." + }, + { + "span_id": "span-1155b3befb7908f7c3de977c", + "text": "- Practical understanding of RAG: embeddings, chunking, vector search, reranking, basic retrieval evaluation." + }, + { + "span_id": "span-c6e390d369c959acd7b87a11", + "text": "- Comfort with Git, Docker, basic cloud deployment, Linux command line." + }, + { + "span_id": "span-c798e0639fd338f8544e8fb6", + "text": "- Habit of evaluating LLM outputs rather than trusting them: written test cases, golden sets, regression checks." + }, + { + "span_id": "span-9668227867faeb304be94b8f", + "text": "- Daily use of agentic coding tools (Claude Code, Codex, Cursor) for real work, with careful inspection of what they produce." + }, + { + "span_id": "span-808b30e6c3502990be723b01", + "text": "- Strong written communication." + }, + { + "span_id": "span-030b15be93d9171c020398fe", + "text": "Clear pull request (PR) descriptions, clear questions, clear documentation." + }, + { + "span_id": "span-e3912f01aa228ac620916464", + "text": "06 Useful but not required" + }, + { + "span_id": "span-47d09037bfe36fce7a1c1a89", + "text": "- Familiarity with LangChain, LangGraph, LlamaIndex, LangSmith, Langfuse, Braintrust, MCP servers, or OpenAI Agents SDK." + }, + { + "span_id": "span-7146fd22469013140db6c2ea", + "text": "- Vector databases such as Pinecone, Weaviate, Qdrant, Chroma, or pgvector." + }, + { + "span_id": "span-8b3d66c870c7f9ce5067fb6f", + "text": "- TypeScript, React, Next.js, Postgres, or Supabase." + }, + { + "span_id": "span-0fa1a81a767ba4af4e931d4e", + "text": "- Domain interest in finance, investment, operations, sales, or customer support." + }, + { + "span_id": "span-2f45cfbd322f4277f38b2b4a", + "text": "- Experience with fine-tuning, distillation, or open-weight models." + }, + { + "span_id": "span-3f2835b11f4f3d1246dee51c", + "text": "07 What you will learn" + } + ] + }, + { + "chunk_id": "tai-ba01d70e2987306bc3f567c1", + "text": "Junior AI Engineer\nnot demos. - How to debug agent failures, fix retrieval quality issues, and harden systems for real use. - How to evaluate models, prompts, and pipelines with discipline rather than vibes. - How to use agentic coding tools at a senior level: planning, supervision, intelligent rejection. - How AI engineering translates into real client value, observed through the senior engineers and strategists you support. 08 Who we are looking for An early-career engineer with strong technical fundamentals, real curiosity about AI engineering as a craft, and the temperament for a high-bar environment. You care about correctness, evaluation, and useful outputs more than impressive vocabulary. You read other people's code carefully, ask sharp questions, ship fast under intelligent supervision of agentic tools, and grow fast. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one. Apply", + "heading": "Junior AI Engineer", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-2261a7b4a50c8ca5368f4327", + "text": "- How to debug agent failures, fix retrieval quality issues, and harden systems for real use." + }, + { + "span_id": "span-0d30c6bf09bb56ae85ee2c50", + "text": "- How to evaluate models, prompts, and pipelines with discipline rather than vibes." + }, + { + "span_id": "span-39e9357c4b98a028b22f1789", + "text": "- How to use agentic coding tools at a senior level: planning, supervision, intelligent rejection." + }, + { + "span_id": "span-823fab9b9a66ada6015b0b70", + "text": "- How AI engineering translates into real client value, observed through the senior engineers and strategists you support." + }, + { + "span_id": "span-633d026ee9c459a60002102c", + "text": "08 Who we are looking for" + }, + { + "span_id": "span-c043a1e734eecf040f8c3dd6", + "text": "An early-career engineer with strong technical fundamentals, real curiosity about AI engineering as a craft, and the temperament for a high-bar environment." + }, + { + "span_id": "span-03f93f21cd748ffa97f9617c", + "text": "You care about correctness, evaluation, and useful outputs more than impressive vocabulary." + }, + { + "span_id": "span-4e105d869715615a3a2c9a52", + "text": "You read other people's code carefully, ask sharp questions, ship fast under intelligent supervision of agentic tools, and grow fast." + }, + { + "span_id": "span-207001c94e24324be6e79150", + "text": "About Towards AI" + }, + { + "span_id": "span-dc13dd11fc5dea18f3438b5e", + "text": "Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production." + }, + { + "span_id": "span-fbcd02a8a2f5847d9f39bccb", + "text": "Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries." + }, + { + "span_id": "span-7800c6eb2fc3b5b4c2927b94", + "text": "Co-founded by Louie Peters (ex-J.P." + }, + { + "span_id": "span-915831f6334d58c2d9bb985d", + "text": "Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal)." + }, + { + "span_id": "span-98cb5296e01e367f7008fd99", + "text": "Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field." + }, + { + "span_id": "span-0c9b1c4c59deab1786552262", + "text": "In a pod you own a real system end to end, in production, from day one." + } + ] + }, + { + "chunk_id": "tai-9d7e49f0276a0fc5d8ed4989", + "text": "Tell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "heading": "Tell us what you’ve shipped .", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-30ed9ade2afc8cbf790c96e5", + "text": "Concrete write-ups beat resumes." + }, + { + "span_id": "span-2fa959ea9d6112daf89d9a9b", + "text": "We read every application." + }, + { + "span_id": "span-41fd67d98a41484cf2d67f54", + "text": "If we want to talk, you’ll hear from us inside a week." + }, + { + "span_id": "span-ef3dcc51ed0b904b3f13d718", + "text": "Full name" + }, + { + "span_id": "span-b2a9d73c62589e4584d57019", + "text": "Country you’re based in" + }, + { + "span_id": "span-a3a3e0779b33828300775538", + "text": "Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role" + }, + { + "span_id": "span-234c2f8cf539b89732ddc52d", + "text": "LinkedIn profile URL" + }, + { + "span_id": "span-ab660eb1cab10e3fca47319b", + "text": "CV / Resume (PDF preferred)" + }, + { + "span_id": "span-13248c7b94463cf84a12728e", + "text": "Link to your best public work" + }, + { + "span_id": "span-1df3790962decb81758151fd", + "text": "Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines)" + }, + { + "span_id": "span-978ff04566d2d181076d7aa1", + "text": "How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work?" + }, + { + "span_id": "span-63390dea1aa3b74958d0bfc8", + "text": "Submit application arrow_forward" + }, + { + "span_id": "span-ea926193a0af02f6989ffef7", + "text": "Apply for this role Apply for this role arrow_forward" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "cc95a4d62d6c8b31e0756a115068933a0f5cd6e3af5e0490ad99ec808231940d", + "content_hash": "cc95a4d62d6c8b31e0756a115068933a0f5cd6e3af5e0490ad99ec808231940d", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/valuecreation/junior-ai-engineer/" + ], + "evidence_hash": "06ecad1bda6a21a188bcaa4bda55065f4ae716d091fc4d6f4aa68886d684cc23" + }, + { + "discovered_url": "https://towardsai.com/valuecreation/senior-ai-engineer/", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/", + "canonical_url": "https://towardsai.com/valuecreation/senior-ai-engineer/", + "host": "towardsai.com", + "path": "/valuecreation/senior-ai-engineer", + "kind": "career", + "offer_id": null, + "entity_id": "page:towardsai.com/valuecreation/senior-ai-engineer", + "title": "Senior AI Engineer / FDE · Towards AI Careers", + "review_title": "", + "reviewed_summary": "", + "meta_description": "Join Towards AI as a Senior AI Engineer / Forward Deployed Engineer, building production AI systems for PE-backed companies from discovery through deployment.", + "headings": [ + "Senior AI Engineer / Forward Deployed Engineer", + "Tell us what you’ve shipped ." + ], + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Senior AI Engineer / FDE\n\nSenior AI Engineer / Forward Deployed Engineer\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Co-founders 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. Around 95% of enterprise AI experiments never reach production. We hire only top-tier talent to close that gap for clients with serious engagement budgets and real urgency to ship. Six years in: 500,000+ learners since 2019, an O'Reilly book referenced inside Fortune 500s, and a 15-person engineering team shipping live work for clients including J.P. Morgan, Intel, Europol, NYPL, Nviya-Prime, Maoki, and Activeloop. You will own the technical delivery of custom AI systems end-to-end: architecture, build, evaluation, hardening, and production rollout. Typical builds include agentic systems for investment research and due diligence, retrieval-augmented copilots for finance workflows, document analysis pipelines, market mapping tools, internal knowledge assistants, systems integrations, and workflow automations for portfolio operations. Most weeks are heads-down engineering with your pod, shipping systems that real client teams depend on day-to-day. 02 A recent example Nviya-Prime We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech replicating institutional-grade banker reasoning. Three coordinated sub-agent services orchestrate ~800 prompts per client analysis. Event Targeting classifies material signals across economic, geopolitical, regulatory, market, supply-chain, and idiosyncratic domains. Predictive Scanning runs 24/7, scoring live events on materiality x urgency and surfacing prioritised next-best-actions, for example an FX hedge plus interest-rate contract in response to a funding-cost spike. Production stack: agentic retrieval-augmented generation (RAG), hybrid search, contextual retrieval, reranking, model routing, LLM-as-judge evaluation, end-to-end observability. Currently shipping into live demos with global banks across Europe. 03 How the engagement works Two modes, depending on the project. For most engagements you work from our team alongside your engineering pod, with regular technical touchpoints into the client's engineering and data teams. A deployment strategist on each engagement owns the executive client relationship and the bulk of stakeholder management, leaving you to focus on getting the system built right. For our largest multi-quarter engagements, typically with financial services firms, PE houses, or strategic portfolio companies, you go forward-deployed at the client site for kickoff sprints, key build phases, and production rollout. Your direct client counterparts are engineering, data, product, and end-user teams. You should be excited by this mode of working: it offers the fastest feedback loops and the most direct influence on what gets shipped of any client setup we run. 04 What you will own - Lead technical scoping and architecture across client engagements. - Build production LLM applications using OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or similar. - Architect RAG systems: retrieval quality, grounding, source attribution, metadata, hybrid search, reranking, access control. - Build agentic workflows: tool use, structured outputs, orchestration, retries, fallbacks, state, human review. - Design eval frameworks for prompts, retrieval, agents, models, regressions, safety, latency, and cost. - Define guardrails for privacy, governance, prompt injection, sensitive tool use, and hallucination risk. - Deploy and monitor with Docker, cloud infrastructure, continuous integration and deployment (CI/CD), logging, and observability. - Pair with client engineering and data teams on integrations, deployment, and technical handover. - Mentor junior engineers; hold the technical quality bar across delivery. - Ship reusable firm assets: reference architectures, prompt libraries, eval harnesses, agent skills, implementation playbooks. 05 AI-native working style Daily, expert use of Claude Code, Codex, Cursor, or similar agentic tools is non-negotiable. Planning, codebase exploration, implementation, testing, refactoring, documentation, research, debugging: all of it. Engineers who already work this way outpace traditional engineers by a wide margin, and that gap is the bar for this role. The standard is intelligent supervision, not blind delegation. You give the agent the right context, constrain the task, inspect the diff, run the tests, catch weak assumptions, and decide when to take the keyboard back. You ship at speed but never ship what you have not understood. You should also actively design reusable AI infrastructure for the firm: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, and workflow templates. Our strongest engineers build the tooling that makes the rest of the team faster. 06 Required qualifications - 5+ years professional software engineering experience. - 2+ years building LLM applications, RAG systems, AI agents, or AI workflow automation in production. - Shipped at least two AI systems beyond simple chat: domain-specific RAG, multi-tool agents, document analysis, internal copilots, or comparable. - Strong Python, plus production experience with application programming interfaces (APIs), integrations, and data-heavy systems. - Hands-on with OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or comparable LLM APIs. - Real RAG depth: embeddings, chunking, indexing, metadata, hybrid search, reranking, retrieval evaluation, grounding. - Real agent depth: tool calling, structured outputs, orchestration, state, retries, safe execution boundaries. - Strong eval discipline: golden sets, regression tests, human review loops, considered use of LLM-as-judge, release gates, production monitoring. - Production judgment on latency, cost, scalability, reliability, privacy, security, and data governance. - Comfort with Docker, Git, CI/CD, cloud deployment, databases, logging, monitoring. - Daily, expert use of agentic coding tools (Claude Code, Codex, Cursor) for real engineering, with careful inspection of what they produce. - Strong technical communication. You can explain architecture, trade-offs, and risk to engineering counterparts and to your own pod, and hold your own in a senior client conversation when the strategist hands you the floor. 07 Useful but not required - Prior work with investors, PE firms, venture funds, family offices, or portfolio companies. - AI experience in investment research, due diligence, finance, sales, customer support, reporting, or knowledge management. - TypeScript, React, Next.js, Postgres, pgvector, Pinecone, Weaviate, or Qdrant. - LangChain, LangGraph, LlamaIndex, LangSmith, Langfuse, Braintrust, OpenTelemetry, MCP servers, OpenAI Agents SDK, or Claude Skills. - Solutions engineering, customer engineering, founder-led product, or enterprise implementation backgrounds. - Fine-tuning, distillation, or open-weight model experience. 08 Who we are looking for A builder and a field engineer. You take an ambiguous technical problem, identify the constraint that matters, and design a system that gets used rather than admired. You move between architecture and code without losing either. You know when an agent is the right shape, when deterministic code is better, when retrieval is the bottleneck, and when a proposed scope is the wrong one. You are AI-native enough that work without these tools feels broken. You should be able to describe a system you have built in concrete terms: problem, architecture, model choices, retrieval design, evals, failure modes, cost profile, rollout plan, and what you would change next time. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one. Apply\n\nTell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "links": [ + { + "text": "Student Academy", + "url": "https://towardsai.com/academy/" + }, + { + "text": "The Book", + "url": "https://towardsai.com/book" + }, + { + "text": "Publication", + "url": "https://pub.towardsai.net/" + }, + { + "text": "Newsletter", + "url": "https://newsletter.towardsai.net/" + }, + { + "text": "AI Jobs", + "url": "https://jobs.towardsai.net/" + }, + { + "text": "What’s AI YouTube", + "url": "https://www.youtube.com/@WhatsAI" + }, + { + "text": "Enterprise Academy", + "url": "https://towardsai.com/enterpriseenablement/" + }, + { + "text": "Value Creation Consulting", + "url": "https://towardsai.com/valuecreation/" + }, + { + "text": "Case studies", + "url": "https://towardsai.com/" + }, + { + "text": "Community arrow_forward", + "url": "https://discord.gg/learnaitogether" + }, + { + "text": "Junior AI Engineer", + "url": "https://towardsai.com/valuecreation/junior-ai-engineer/" + }, + { + "text": "Senior AI Engineer / FDE", + "url": "https://towardsai.com/valuecreation/senior-ai-engineer/" + }, + { + "text": "Deployment Strategist", + "url": "https://towardsai.com/valuecreation/deployment-strategist/" + }, + { + "text": "Explore courses", + "url": "https://towardsai.com/academy/" + }, + { + "text": "Talk to us", + "url": "https://towardsai.com/" + }, + { + "text": "Back to Towards AI Back to Towards AI", + "url": "https://towardsai.com/" + } + ], + "chunks": [ + { + "chunk_id": "tai-1124ca4dd6f9ade9bbcb2bad", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Explore courses Talk to us Towards AI / Careers - we’re hiring / Senior AI Engineer / FDE", + "heading": "Overview", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-e0e2aeee19593935a56f102f", + "text": "For Learners keyboard_arrow_down Student Academy The Book Publication Newsletter AI Jobs What’s AI YouTube" + }, + { + "span_id": "span-85e5ecc6982540b5430f04d0", + "text": "For Enterprise keyboard_arrow_down Enterprise Academy Value Creation Consulting Case studies" + }, + { + "span_id": "span-4a618e9f574107ffe74644b0", + "text": "Community arrow_forward Careers - we’re hiring keyboard_arrow_down Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist" + }, + { + "span_id": "span-afafc51b2e9e52a615d5cfde", + "text": "Explore courses Talk to us" + }, + { + "span_id": "span-55e4e5beeafd0106d27abb19", + "text": "Towards AI / Careers - we’re hiring / Senior AI Engineer / FDE" + } + ] + }, + { + "chunk_id": "tai-f97707ed0b92e85ddcb8caf9", + "text": "Senior AI Engineer / Forward Deployed Engineer\nWe bring AI inside organisations. We build the systems. We train the people. Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI Team AI Engineering / Value Creation Location London or Hybrid Apply by Rolling Reports to Co-founders 01 The opportunity We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong. Around 95% of enterprise AI experiments never reach production. We hire only top-tier talent to close that gap for clients with serious engagement budgets and real urgency to ship. Six years in: 500,000+ learners since 2019, an O'Reilly book referenced inside Fortune 500s, and a 15-person engineering team shipping live work for clients including J.P. Morgan, Intel, Europol, NYPL, Nviya-Prime, Maoki, and Activeloop. You will own the technical delivery of custom AI systems end-to-end: architecture, build, evaluation, hardening, and production rollout. Typical builds include agentic systems for investment research and due diligence, retrieval-augmented copilots for finance workflows, document analysis pipelines, market mapping tools, internal knowledge assistants, systems integrations, and workflow automations for portfolio operations. Most weeks are heads-down engineering with your pod, shipping systems that real client teams depend on day-to-day. 02 A recent example Nviya-Prime We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech replicating institutional-grade banker reasoning. Three coordinated sub-agent services orchestrate ~800 prompts per client analysis. Event Targeting classifies material signals across economic, geopolitical,", + "heading": "Senior AI Engineer / Forward Deployed Engineer", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-62bf6fff211fdea79a0aac02", + "text": "We bring AI inside organisations." + }, + { + "span_id": "span-b2c28b765dd3c2a510d8976e", + "text": "We build the systems." + }, + { + "span_id": "span-2dc7668f7c41040b25551081", + "text": "We train the people." + }, + { + "span_id": "span-84cec96d3abd8b5c6ffff2a6", + "text": "Apply for this role Apply for this role arrow_forward Back to Towards AI Back to Towards AI" + }, + { + "span_id": "span-f9a7506dfce8e27a28ccc842", + "text": "AI Engineering / Value Creation" + }, + { + "span_id": "span-a87d7f850e064b4a967eabde", + "text": "London or Hybrid" + }, + { + "span_id": "span-7ac98c4bc2b3cca983523f07", + "text": "Apply by" + }, + { + "span_id": "span-cbc977379a4d065a1c487d91", + "text": "Reports to" + }, + { + "span_id": "span-c1b37737acb44ad2e944b66d", + "text": "Co-founders" + }, + { + "span_id": "span-483497084214bde2e9b28a2a", + "text": "01 The opportunity" + }, + { + "span_id": "span-50bfe677323ad75aa93f1659", + "text": "We are building a premium AI consultancy targeting an underserved whitespace: transforming small-to-mid-cap portfolio companies and finance firms that cannot afford to get AI wrong." + }, + { + "span_id": "span-8fac821e322a11af624c58e3", + "text": "Around 95% of enterprise AI experiments never reach production." + }, + { + "span_id": "span-e7210dfdacb23247313cb8f5", + "text": "We hire only top-tier talent to close that gap for clients with serious engagement budgets and real urgency to ship." + }, + { + "span_id": "span-2022ba16b4a9908f8ebe022e", + "text": "Six years in: 500,000+ learners since 2019, an O'Reilly book referenced inside Fortune 500s, and a 15-person engineering team shipping live work for clients including J.P." + }, + { + "span_id": "span-8ed2debd0bf527563ce3d2ff", + "text": "Morgan, Intel, Europol, NYPL, Nviya-Prime, Maoki, and Activeloop." + }, + { + "span_id": "span-cbf21f9cd713cab04800878c", + "text": "You will own the technical delivery of custom AI systems end-to-end: architecture, build, evaluation, hardening, and production rollout." + }, + { + "span_id": "span-3c2b570246cc067a69830f15", + "text": "Typical builds include agentic systems for investment research and due diligence, retrieval-augmented copilots for finance workflows, document analysis pipelines, market mapping tools, internal knowledge assistants, systems integrations, and workflow automations for portfolio operations." + }, + { + "span_id": "span-ec0ec2b8193032784290cc00", + "text": "Most weeks are heads-down engineering with your pod, shipping systems that real client teams depend on day-to-day." + }, + { + "span_id": "span-09e0d2bf8360218d10e57cb0", + "text": "02 A recent example" + }, + { + "span_id": "span-b33e5e6f2bd4d652d6c23087", + "text": "Nviya-Prime" + }, + { + "span_id": "span-fa0310c8427c4f5b0f856f03", + "text": "We architected the Agentic Intelligence Engine for Nviya-Prime, a B2B FinTech replicating institutional-grade banker reasoning." + }, + { + "span_id": "span-813d83af74e93cfddb91808a", + "text": "Three coordinated sub-agent services orchestrate ~800 prompts per client analysis." + } + ] + }, + { + "chunk_id": "tai-9ab97e8a120121773a0d2149", + "text": "Senior AI Engineer / Forward Deployed Engineer\nregulatory, market, supply-chain, and idiosyncratic domains. Predictive Scanning runs 24/7, scoring live events on materiality x urgency and surfacing prioritised next-best-actions, for example an FX hedge plus interest-rate contract in response to a funding-cost spike. Production stack: agentic retrieval-augmented generation (RAG), hybrid search, contextual retrieval, reranking, model routing, LLM-as-judge evaluation, end-to-end observability. Currently shipping into live demos with global banks across Europe. 03 How the engagement works Two modes, depending on the project. For most engagements you work from our team alongside your engineering pod, with regular technical touchpoints into the client's engineering and data teams. A deployment strategist on each engagement owns the executive client relationship and the bulk of stakeholder management, leaving you to focus on getting the system built right. For our largest multi-quarter engagements, typically with financial services firms, PE houses, or strategic portfolio companies, you go forward-deployed at the client site for kickoff sprints, key build phases, and production rollout. Your direct client counterparts are engineering, data, product, and end-user teams. You should be excited by this mode of working: it offers the fastest feedback loops and the most direct influence on what gets shipped of any client setup we run. 04 What you will own - Lead technical scoping and architecture across client engagements. - Build production LLM applications using OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or similar. - Architect RAG systems: retrieval quality, grounding, source attribution, metadata, hybrid search, reranking, access control. - Build agentic workflows: tool use,", + "heading": "Senior AI Engineer / Forward Deployed Engineer", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-50cb19e21ed4aa7855732837", + "text": "Predictive Scanning runs 24/7, scoring live events on materiality x urgency and surfacing prioritised next-best-actions, for example an FX hedge plus interest-rate contract in response to a funding-cost spike." + }, + { + "span_id": "span-a4dcf2cce33732d4e56f496d", + "text": "Production stack: agentic retrieval-augmented generation (RAG), hybrid search, contextual retrieval, reranking, model routing, LLM-as-judge evaluation, end-to-end observability." + }, + { + "span_id": "span-378d0e011c6d998f36872e7e", + "text": "Currently shipping into live demos with global banks across Europe." + }, + { + "span_id": "span-0f140afd9fb8b43426a602c5", + "text": "03 How the engagement works" + }, + { + "span_id": "span-4d74aaed215cfb944dd712c5", + "text": "Two modes, depending on the project." + }, + { + "span_id": "span-0964f29ebfd13ca6e53663d2", + "text": "For most engagements you work from our team alongside your engineering pod, with regular technical touchpoints into the client's engineering and data teams." + }, + { + "span_id": "span-3047c684f7357585c4a48fe5", + "text": "A deployment strategist on each engagement owns the executive client relationship and the bulk of stakeholder management, leaving you to focus on getting the system built right." + }, + { + "span_id": "span-80915906ecc3d42b2cf1218d", + "text": "For our largest multi-quarter engagements, typically with financial services firms, PE houses, or strategic portfolio companies, you go forward-deployed at the client site for kickoff sprints, key build phases, and production rollout." + }, + { + "span_id": "span-c8087834034bb04f2b98fd6d", + "text": "Your direct client counterparts are engineering, data, product, and end-user teams." + }, + { + "span_id": "span-e7cf70d8d7109cc6ab628355", + "text": "You should be excited by this mode of working: it offers the fastest feedback loops and the most direct influence on what gets shipped of any client setup we run." + }, + { + "span_id": "span-48bc65fcb3b2eba710bd777d", + "text": "04 What you will own" + }, + { + "span_id": "span-a44b7550d189885bcc0b8a4a", + "text": "- Lead technical scoping and architecture across client engagements." + }, + { + "span_id": "span-b15e8e610cad075773234602", + "text": "- Build production LLM applications using OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or similar." + }, + { + "span_id": "span-269c98ae369852d6de820b01", + "text": "- Architect RAG systems: retrieval quality, grounding, source attribution, metadata, hybrid search, reranking, access control." + } + ] + }, + { + "chunk_id": "tai-c7867c0b36ca353856b610bf", + "text": "Senior AI Engineer / Forward Deployed Engineer\nstructured outputs, orchestration, retries, fallbacks, state, human review. - Design eval frameworks for prompts, retrieval, agents, models, regressions, safety, latency, and cost. - Define guardrails for privacy, governance, prompt injection, sensitive tool use, and hallucination risk. - Deploy and monitor with Docker, cloud infrastructure, continuous integration and deployment (CI/CD), logging, and observability. - Pair with client engineering and data teams on integrations, deployment, and technical handover. - Mentor junior engineers; hold the technical quality bar across delivery. - Ship reusable firm assets: reference architectures, prompt libraries, eval harnesses, agent skills, implementation playbooks. 05 AI-native working style Daily, expert use of Claude Code, Codex, Cursor, or similar agentic tools is non-negotiable. Planning, codebase exploration, implementation, testing, refactoring, documentation, research, debugging: all of it. Engineers who already work this way outpace traditional engineers by a wide margin, and that gap is the bar for this role. The standard is intelligent supervision, not blind delegation. You give the agent the right context, constrain the task, inspect the diff, run the tests, catch weak assumptions, and decide when to take the keyboard back. You ship at speed but never ship what you have not understood. You should also actively design reusable AI infrastructure for the firm: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, and workflow templates. Our strongest engineers build the tooling that makes the rest of the team faster. 06 Required qualifications - 5+ years professional software engineering experience. - 2+ years building LLM", + "heading": "Senior AI Engineer / Forward Deployed Engineer", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-40ae2e94d4375566c555b57d", + "text": "- Design eval frameworks for prompts, retrieval, agents, models, regressions, safety, latency, and cost." + }, + { + "span_id": "span-bd49c0d46feeb37c688f5cac", + "text": "- Define guardrails for privacy, governance, prompt injection, sensitive tool use, and hallucination risk." + }, + { + "span_id": "span-f08c63c70db9f754d553c791", + "text": "- Deploy and monitor with Docker, cloud infrastructure, continuous integration and deployment (CI/CD), logging, and observability." + }, + { + "span_id": "span-e3fbcb24b3a00613be281721", + "text": "- Pair with client engineering and data teams on integrations, deployment, and technical handover." + }, + { + "span_id": "span-0a46b196eaac2d26e1ac4417", + "text": "- Mentor junior engineers; hold the technical quality bar across delivery." + }, + { + "span_id": "span-f2b947fa0671f1877ee092d4", + "text": "- Ship reusable firm assets: reference architectures, prompt libraries, eval harnesses, agent skills, implementation playbooks." + }, + { + "span_id": "span-626aad1099d7da52536e8161", + "text": "05 AI-native working style" + }, + { + "span_id": "span-b841b1deb18a7f64e8ad4519", + "text": "Daily, expert use of Claude Code, Codex, Cursor, or similar agentic tools is non-negotiable." + }, + { + "span_id": "span-e4045ac1c1a4cdad8a32214a", + "text": "Planning, codebase exploration, implementation, testing, refactoring, documentation, research, debugging: all of it." + }, + { + "span_id": "span-828cd9b54aa27b700ff2a44c", + "text": "Engineers who already work this way outpace traditional engineers by a wide margin, and that gap is the bar for this role." + }, + { + "span_id": "span-d6dc52197b1a89be7605a840", + "text": "The standard is intelligent supervision, not blind delegation." + }, + { + "span_id": "span-9ff939160050451dedc73a18", + "text": "You give the agent the right context, constrain the task, inspect the diff, run the tests, catch weak assumptions, and decide when to take the keyboard back." + }, + { + "span_id": "span-90115fe23402ceb067fa1168", + "text": "You ship at speed but never ship what you have not understood." + }, + { + "span_id": "span-9747b55541be329bc717e832", + "text": "You should also actively design reusable AI infrastructure for the firm: Claude Skills, Model Context Protocol (MCP) servers, sub-agents, eval harnesses, prompt patterns, and workflow templates." + }, + { + "span_id": "span-05c59e36ff59356f1e7b79e1", + "text": "Our strongest engineers build the tooling that makes the rest of the team faster." + }, + { + "span_id": "span-383e46a11ad62a79bc46591a", + "text": "06 Required qualifications" + }, + { + "span_id": "span-e84761bff726e8e955b33a4c", + "text": "- 5+ years professional software engineering experience." + } + ] + }, + { + "chunk_id": "tai-3de95c89109500f094714ce0", + "text": "Senior AI Engineer / Forward Deployed Engineer\napplications, RAG systems, AI agents, or AI workflow automation in production. - Shipped at least two AI systems beyond simple chat: domain-specific RAG, multi-tool agents, document analysis, internal copilots, or comparable. - Strong Python, plus production experience with application programming interfaces (APIs), integrations, and data-heavy systems. - Hands-on with OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or comparable LLM APIs. - Real RAG depth: embeddings, chunking, indexing, metadata, hybrid search, reranking, retrieval evaluation, grounding. - Real agent depth: tool calling, structured outputs, orchestration, state, retries, safe execution boundaries. - Strong eval discipline: golden sets, regression tests, human review loops, considered use of LLM-as-judge, release gates, production monitoring. - Production judgment on latency, cost, scalability, reliability, privacy, security, and data governance. - Comfort with Docker, Git, CI/CD, cloud deployment, databases, logging, monitoring. - Daily, expert use of agentic coding tools (Claude Code, Codex, Cursor) for real engineering, with careful inspection of what they produce. - Strong technical communication. You can explain architecture, trade-offs, and risk to engineering counterparts and to your own pod, and hold your own in a senior client conversation when the strategist hands you the floor. 07 Useful but not required - Prior work with investors, PE firms, venture funds, family offices, or portfolio companies. - AI experience in investment research, due diligence, finance, sales, customer support, reporting, or knowledge management. - TypeScript, React, Next.js, Postgres, pgvector, Pinecone, Weaviate, or Qdrant. - LangChain, LangGraph, LlamaIndex, LangSmith,", + "heading": "Senior AI Engineer / Forward Deployed Engineer", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-0ba2abaf9e2f4625141e43da", + "text": "- Shipped at least two AI systems beyond simple chat: domain-specific RAG, multi-tool agents, document analysis, internal copilots, or comparable." + }, + { + "span_id": "span-cbbf21576c2c006ed75c8e71", + "text": "- Strong Python, plus production experience with application programming interfaces (APIs), integrations, and data-heavy systems." + }, + { + "span_id": "span-3b4d6585393a83be911937bd", + "text": "- Hands-on with OpenAI, Anthropic, Gemini, Bedrock, Vertex AI, or comparable LLM APIs." + }, + { + "span_id": "span-42a1e41a9bfe79e2dc75ed03", + "text": "- Real RAG depth: embeddings, chunking, indexing, metadata, hybrid search, reranking, retrieval evaluation, grounding." + }, + { + "span_id": "span-e4408b05f8ee69f1353e3cf0", + "text": "- Real agent depth: tool calling, structured outputs, orchestration, state, retries, safe execution boundaries." + }, + { + "span_id": "span-bd5fd90b006848ba62dd91eb", + "text": "- Strong eval discipline: golden sets, regression tests, human review loops, considered use of LLM-as-judge, release gates, production monitoring." + }, + { + "span_id": "span-138e7bf16e0e006f13aa9688", + "text": "- Production judgment on latency, cost, scalability, reliability, privacy, security, and data governance." + }, + { + "span_id": "span-98d40310123c83e9239df0fd", + "text": "- Comfort with Docker, Git, CI/CD, cloud deployment, databases, logging, monitoring." + }, + { + "span_id": "span-75c50235615bbfd141bf5688", + "text": "- Daily, expert use of agentic coding tools (Claude Code, Codex, Cursor) for real engineering, with careful inspection of what they produce." + }, + { + "span_id": "span-b87c4a064a95f8e760b3a8c0", + "text": "- Strong technical communication." + }, + { + "span_id": "span-aee95d4abfa17377a71469ba", + "text": "You can explain architecture, trade-offs, and risk to engineering counterparts and to your own pod, and hold your own in a senior client conversation when the strategist hands you the floor." + }, + { + "span_id": "span-242a4db579b0c5f57617258b", + "text": "07 Useful but not required" + }, + { + "span_id": "span-f567ba733ab8311b737de814", + "text": "- Prior work with investors, PE firms, venture funds, family offices, or portfolio companies." + }, + { + "span_id": "span-820f377bb4e29b2bb0c57efa", + "text": "- AI experience in investment research, due diligence, finance, sales, customer support, reporting, or knowledge management." + }, + { + "span_id": "span-9b60ee3f88c4c951c6a8393c", + "text": "- TypeScript, React, Next.js, Postgres, pgvector, Pinecone, Weaviate, or Qdrant." + } + ] + }, + { + "chunk_id": "tai-4c69d73eb7c43e03f3874439", + "text": "Senior AI Engineer / Forward Deployed Engineer\nLangfuse, Braintrust, OpenTelemetry, MCP servers, OpenAI Agents SDK, or Claude Skills. - Solutions engineering, customer engineering, founder-led product, or enterprise implementation backgrounds. - Fine-tuning, distillation, or open-weight model experience. 08 Who we are looking for A builder and a field engineer. You take an ambiguous technical problem, identify the constraint that matters, and design a system that gets used rather than admired. You move between architecture and code without losing either. You know when an agent is the right shape, when deterministic code is better, when retrieval is the bottleneck, and when a proposed scope is the wrong one. You are AI-native enough that work without these tools feels broken. You should be able to describe a system you have built in concrete terms: problem, architecture, model choices, retrieval design, evals, failure modes, cost profile, rollout plan, and what you would change next time. About Towards AI Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production. Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries. Co-founded by Louie Peters (ex-J.P. Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal). Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field. In a pod you own a real system end to end, in production, from day one.", + "heading": "Senior AI Engineer / Forward Deployed Engineer", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-6abcfa1e9d29b2aa5c3cc821", + "text": "- Solutions engineering, customer engineering, founder-led product, or enterprise implementation backgrounds." + }, + { + "span_id": "span-d425dd88532dfd59eedf696c", + "text": "- Fine-tuning, distillation, or open-weight model experience." + }, + { + "span_id": "span-27a6ebcee216a13ef8f20449", + "text": "08 Who we are looking for" + }, + { + "span_id": "span-dbeff6e96a1636833b36e84b", + "text": "A builder and a field engineer." + }, + { + "span_id": "span-44823fd0f083da9526670552", + "text": "You take an ambiguous technical problem, identify the constraint that matters, and design a system that gets used rather than admired." + }, + { + "span_id": "span-b3ec14736db46f6389911e75", + "text": "You move between architecture and code without losing either." + }, + { + "span_id": "span-7475eaa17b73e0f9ab8dabf3", + "text": "You know when an agent is the right shape, when deterministic code is better, when retrieval is the bottleneck, and when a proposed scope is the wrong one." + }, + { + "span_id": "span-a81bbc5fc34da28ba02f9dac", + "text": "You are AI-native enough that work without these tools feels broken." + }, + { + "span_id": "span-da18607611472e3304d4bcf2", + "text": "You should be able to describe a system you have built in concrete terms: problem, architecture, model choices, retrieval design, evals, failure modes, cost profile, rollout plan, and what you would change next time." + }, + { + "span_id": "span-d32a02b47a187bad05437884", + "text": "About Towards AI" + }, + { + "span_id": "span-29aef7bb142152f02b773b87", + "text": "Towards AI is the AI deployment and education firm, founded in 2019: 500,000+ AI practitioners taught, 200,000+ newsletter subscribers, 100,000+ community members, and 10,000+ copies sold of our O'Reilly book Building LLMs for Production." + }, + { + "span_id": "span-d267c75868fa16c40419d340", + "text": "Our Towards AI Deployment practice runs a specialist team of 15+ AI engineers focused on investment firms, PE, portfolio companies, and finance firms in regulated industries." + }, + { + "span_id": "span-eb484677e12826d047dcb07d", + "text": "Co-founded by Louie Peters (ex-J.P." + }, + { + "span_id": "span-3e58c49879edb35fd42a9918", + "text": "Morgan VP, credit research) and Louis-Francois Bouchard (ex-Mila, Polytechnique Montreal)." + }, + { + "span_id": "span-49a1647283812467fd89b2ff", + "text": "Our engineers grow up through our own community and Learning programmes into Deployment pods, so you would work with people who stay at the forefront of the field." + }, + { + "span_id": "span-1a5b8a02a87ec10b6c83c0ce", + "text": "In a pod you own a real system end to end, in production, from day one." + } + ] + }, + { + "chunk_id": "tai-27d8777f4b1eb90eb020d1fa", + "text": "Senior AI Engineer / Forward Deployed Engineer\nApply\n\nTell us what you’ve shipped .\nConcrete write-ups beat resumes. We read every application. If we want to talk, you’ll hear from us inside a week. Full name Email Country you’re based in Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role LinkedIn profile URL CV / Resume (PDF preferred) Link to your best public work Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines) How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work? Submit application arrow_forward Apply for this role Apply for this role arrow_forward", + "heading": "Tell us what you’ve shipped .", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-b139597f4650e50743b9c903", + "text": "Senior AI Engineer / Forward Deployed Engineer" + }, + { + "span_id": "span-3b103524a9169b76d417d6be", + "text": "Concrete write-ups beat resumes." + }, + { + "span_id": "span-e90dc1ce412613a9e4974bc6", + "text": "We read every application." + }, + { + "span_id": "span-bb5ffe3f1f9796bc2136eeab", + "text": "If we want to talk, you’ll hear from us inside a week." + }, + { + "span_id": "span-f487496af9d0a7a1b43f17eb", + "text": "Full name" + }, + { + "span_id": "span-1261289fb5e1a8bbea0e5299", + "text": "Country you’re based in" + }, + { + "span_id": "span-13c691f618b3772a07361510", + "text": "Role you’re applying for Junior AI Engineer Senior AI Engineer / FDE Deployment Strategist Other / future role" + }, + { + "span_id": "span-5b6cc75c12e77250e22dae68", + "text": "LinkedIn profile URL" + }, + { + "span_id": "span-df8f23e406f5c15d53b91c77", + "text": "CV / Resume (PDF preferred)" + }, + { + "span_id": "span-35cc74df87bda4cc08a2f0f2", + "text": "Link to your best public work" + }, + { + "span_id": "span-8343edd0fc20a63a25530c26", + "text": "Describe the most useful AI-relevant work or outcome you have shipped or driven (2-5 lines)" + }, + { + "span_id": "span-07505d8925662be6bfa5f372", + "text": "How do you use AI tools (Claude, ChatGPT, Claude Code, Codex, Cursor) in your daily work?" + }, + { + "span_id": "span-1568432b9b7abd5cd4629593", + "text": "Submit application arrow_forward" + }, + { + "span_id": "span-a5eccca43ce8e55c6e096686", + "text": "Apply for this role Apply for this role arrow_forward" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "0a51ecf876934e6d954cc15b332456fa96b68765dcb1e5d39a13c47ef0113930", + "content_hash": "0a51ecf876934e6d954cc15b332456fa96b68765dcb1e5d39a13c47ef0113930", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/valuecreation/senior-ai-engineer/" + ], + "evidence_hash": "2c01e9d0a4620b2951f8064dbb05d81674f89d74ca75c3470ff52f1ee14bafa3" + }, + { + "discovered_url": "https://towardsai.com/theaitastegap/", + "url": "https://towardsai.com/theaitastegap/", + "canonical_url": "https://towardsai.com/theaitastegap/", + "host": "towardsai.com", + "path": "/theaitastegap", + "kind": "page", + "offer_id": null, + "entity_id": "page:towardsai.com/theaitastegap", + "title": "The Taste Gap: Why AI work that lands is rare", + "review_title": "", + "reviewed_summary": "", + "meta_description": "The gap between AI slop and AI work that lands is a problem of execution, not capability. How the teams pulling ahead put expert judgment back into every step of the AI workflow.", + "headings": [ + "AI is everywhere. AI work that lands is rare. Why?", + "Two questions, four kinds of operator.", + "Has taste, skips it", + "Has taste, uses it", + "No taste, no AI", + "No taste, uses AI", + "Generic prompts tend toward generic output.", + "For most people, average is a promotion.", + "AI slop is repetition at scale.", + "The expert who hands the keys over.", + "The expert who directs the loop.", + "Vibe coding makes everyone look like a builder.", + "The audience asks the wrong question.", + "AI detectors solve the wrong puzzle.", + "Even so, scrub the obvious tells.", + "Surface signals that weaken trust", + "World-class hidden human talent can be revealed by AI.", + "Companies move through four stages of AI adoption.", + "The next jump is from general tools to custom applications.", + "What custom adds", + "AI as an amplifier of taste." + ], + "text": "The Taste Gap 1 The question\n\nAI is everywhere. AI work that lands is rare. Why?\nThe gap between AI slop and AI work has become a problem of execution, not capability. AI can be used to increase the quality of your work, but for many it is reducing it. To produce a great output, and not just a passable one, AI needs your help. The teams pulling ahead are putting expert judgment back into the workflow at every step. Some of that work requires structured AI training. Some happens in building custom AI systems. Towards AI Research / 2026 2 The map\n\nTwo questions, four kinds of operator.\nTwo questions split the field. Does the operator have taste at the task? Do they bring that taste into the AI workflow? The upper-right box is where the leapfrog is most likely. The expert keeps judgment in the loop, rejects the first draft, and turns the model into a tool for range, speed, and finish. Most AI debate confuses these boxes, and most companies do too. Many rollouts accidentally move experts into the trap: the model gets the task, but the expert's judgment never reaches the work. HAS TASTE AT TASK →\n\nHas taste, skips it\nThe expert who outsources judgment to the model. Output drifts back toward average. THE TRAP\n\nHas taste, uses it\nDirection at every step. Work gains range, speed, and finish without losing judgment. THE WIN\n\nNo taste, no AI\nBelow a useful baseline and stays there. The pre-AI default for non-experts at this task. UNCHANGED\n\nNo taste, uses AI\nLifted to average. A real and useful gain. This is most users today. THE LIFT INJECTS TASTE INTO AI WORKFLOW → Fig 02 / The operating model behind good and bad AI work. 3 The baseline\n\nGeneric prompts tend toward generic output.\nThe model was trained on a large body of text, pictures, and code. Without direction, it tends to return the patterns it has seen most often. It scores many possible next words and turns those scores into probabilities. In unguided use, familiar openings, safe structures, and obvious punchlines often sit near the top. That default behaviour matters. Interesting work usually begins when a person changes the path with taste, constraints, examples, rejected drafts, and sharper standards. Run the prompt cold and you get the median answer, polished into clean sentences. Fig 03 / The bell curve shows how skill is distributed. The AI baseline is the vertical line at the mean. 4 The lift\n\nFor most people, average is a promotion.\nAt many specific tasks, people sit below a useful baseline. A support agent writes a clearer refund note. A founder turns messy notes into a tolerable investor update. A junior analyst gets the first SQL query working instead of staring at a blank editor. This is real value. The model can supply structure, grammar, syntax, and a first pass at the thing the person could not yet make alone. This is the easiest part of the AI story to defend, and the part most often dismissed. Baseline competence matters, even when the result is useful rather than distinctive. Fig 04 / Each dot is one person. Below-baseline people rise vertically to the average. 5 The flood\n\nAI slop is repetition at scale.\nA writing quirk can be charming when one person owns it. A familiar cadence, tidy transition, stock joke, or neat three-part structure may pass unnoticed in a single piece. The irritation starts when a million people ship the same move. The phrase stops reading like style and starts reading like residue. When sameness is everywhere, average output becomes brand drag, review burden, and operational noise. Fig 05 / One repeated move can be harmless. At scale, the pattern becomes the problem. 6 The trap\n\nThe expert who hands the keys over.\nAn expert has real taste built over years. They open the model, type a generic prompt, and ship whatever comes back. The output may look polished, but the expert is no longer present in the work. Their edge came from judgment. The prompt removed it. What used to be their value is now a setting they did not change. Skill that does not enter the workflow does not enter the work. Fig 06 / The expert who skips the taste step ends up exactly where everyone else ends up. 7 The leapfrog\n\nThe expert who directs the loop.\nA smaller group does the harder thing. They bring decades of career experience: real customer conversations, hard-won edge cases, and industry memory that was never cleanly captured in the model's training set. Then they reject the first draft, ask for alternatives, compare them, name what is wrong, and drive it into a second shape. AI has jagged intelligence: superhuman in some narrow moves, weak in places a good operator spots immediately. The leapfrog happens when the model supplies range and speed while the human supplies taste, context, touch, and judgment. The reader cannot always name why, but the work feels chosen rather than produced. Fig 07 / The arrow visibly arcs over the holdout. Same starting skill, very different finish. 8 The vibe gap\n\nVibe coding makes everyone look like a builder.\nA non-engineer can vibe-code a working app in an afternoon. A non-designer can produce a credible landing page in an hour. A non-writer can put a long report on the page by lunch. The speed gains are real and unevenly distributed: people who used to be locked out of building the artifact at all are now shipping demos. For first-draft work, AI compressed the visible part of the job by something like an order of magnitude, and that compression is not going away. The artifact is not the deliverable. A demoable app still has to choose its state model, handle auth and rate limits, survive a flaky network, deal with the empty state and the error state, log enough to debug an outage, and make the small interface decisions that decide whether anyone returns next week. A polished report still needs the source check, the contradicting study the model did not surface, the structural cut, the falsifiable claim, and the line that earns the close. The discount on that work is much smaller than the discount on the first draft. That second layer still belongs to expert hours. The total project is faster than it used to be, but the savings curve is steepest at the start and flattens hard once the surface looks done. A vibe-coded prototype that took three hours can take another forty before it scales as a product. A vibe-written report that took ninety minutes can take another two days before it survives review. The ratio inverts: most of the remaining time sits in the part nobody outside the team can see. That is where the manager trap lives. The person who built the prototype reads it as ninety percent done. The engineer or editor who can name what is missing reads it as forty, and the schedule is being set by the higher number. The same pattern shows up in design comps that fall apart on a single user test, in research output that breaks on the first peer challenge, and in marketing copy that does not survive a legal read. Vibe coding makes everyone look like a builder. The expert is the one who knows what is still missing. 9 The wrong question\n\nThe audience asks the wrong question.\nThe flood of average AI work has bred suspicion of all AI work. Readers, viewers, hiring managers, editors. The reflex is to ask whether a tool was used at all. The reflex has a real cause. Most published AI work today comes from free, weaker models. A smaller share comes from strong models used without judgment in the loop. Either way, the median output is generic, and the suspicion is earned. The trap is letting that suspicion become a tool test rather than a quality test. Agentic coding has already broken that reflex. No serious person sneers anymore because a developer did not type every line by hand. The useful signal is whether the person understood the system, made the right tradeoffs, and owned the result. The same standard should apply outside code. Ask whether human judgment shaped the work. \"Did you use AI?\" Was there human judgment in the loop? Fig 08 / Provenance is the weaker question. Judgment is the stronger one. 10 AI detectors\n\nAI detectors solve the wrong puzzle.\nSet aside the long history of false positives. Assume AI detectors worked perfectly. They would tell you whether tokens look machine-generated. They would not tell you whether a person with judgment shaped the result. The dimension everyone cares about, quality and direction, is the one thing the tool cannot see. Editors, teachers, and hiring managers using detectors are sorting on the wrong axis. DETECTOR VERDICT AI probability 92.4% Tokens flagged 147 / 612 Conclusion LIKELY AI Not measured: Human judgment, taste, structure, originality, point of view, accuracy, purpose. Fig 09 / Confidence on the wrong axis. 11 Practical hygiene\n\nEven so, scrub the obvious tells.\nYou can put real insight and effort into an AI collaboration and still lose the reader at the surface. A few familiar AI signatures can make careful work look careless. Even the strong models leak the same signatures. The tics are model-wide, not a sign of weak models. Careful work that does not get scrubbed reads like the median and gets judged that way. If you want the work judged on its merits, strip the signals that mean nothing about the work and everything about the source.\n\nSurface signals that weaken trust\nEm dashes used as a tic Forced contrast cadence Over-polished stock vocabulary Three bullet conclusions to every section Unnecessary symmetric tricolons Hedge phrases in every paragraph Generic \"in conclusion\" wrap-ups Fig 10 / Surface fixes that buy you a fair reading. 12 The bottleneck unlock\n\nWorld-class hidden human talent can be revealed by AI.\nA great output requires a package of skills; many people have top-tier talent in one area that never reaches an audience because their surrounding skills are subpar. The weakest skill sets the ceiling. A person can be world-class at one thing and remain blocked because the surrounding skills sit below the publishable line. AI can lift the supporting skills above that line, and the talent finally reaches an audience. Rare talent Clears bar Below bar Lifted by AI Songwriter · generational hooks Bottleneck: vocals + production Writes melodies people will hum for years. Cannot sing them. Has no idea how to mix. Without AI Does not ship Melody Lyrics ✕ Vocals ✕ Mix Two failing skills keep the songs in a notebook. With AI Ships full package Melody Lyrics Vocals Mix AI vocals and mix. The hooks finally reach listeners. Operations lead · rare process judgment Bottleneck: tooling + reporting Knows where work gets stuck. Cannot turn that knowledge into a reliable internal tool. Without AI Manual process Process ✕ Data ✕ Tooling ✕ Reports Expert time leaks into cleanup, reminders, and status chasing. With AI Workflow ships Process Data Tooling Reports A custom workflow fills the gaps. Process judgment becomes operational leverage. Fig 11 / The dashed line is the publishable bar. The weakest leg decides if the package ships. 13 The maturity curve\n\nCompanies move through four stages of AI adoption.\nStage zero is familiar work without AI. It is slower, but the accountability chain is clear. People know who made the call and where judgment entered the work. Stage one is shadow AI. People grab free tools, paste work into generic interfaces, and invent private workflows. Quality can dip below the pre-AI line when experts delegate judgment instead of applying it. Risk rises too, from data leakage to brand damage to compliance gaps. Stage two is managed AI. Sanctioned tools, paid licences, policies, and training pull quality back up. The company is using AI on purpose, but much of the work still happens inside generic interfaces. Stage three is embedded AI. Custom workflows are built into core processes, with expert review points, domain rules, retrieval, evals, and audit trails, and the leapfrog from earlier sections is scaled across an organisation. The goal is higher-quality output from humans and AI together. Speed comes for free at every stage. Quality only rises when human judgment lives inside the system rather than arriving as cleanup. Fig 12 / Shadow AI is where risk rises before quality systems catch up. 14 From general to custom\n\nThe next jump is from general tools to custom applications.\nEnterprise adoption is real, but many companies still use the same generic interface for specialized work. That leaves expert judgment outside the system, where it arrives late as correction, cleanup, or rejection. A generic chatbot can answer a question. A vibe-coded app can demo an idea. Neither automatically carries an engineer's taste for states, failure modes, latency, permissions, data shape, review loops, or the small interface decisions that make a solution feel inevitable. The larger value sits in applications built around the work itself: role-specific interfaces, company knowledge, domain rules, human review points, evals, audit trails, and integrations into existing systems. What an engineering team adds that a generic chatbox or quick demo cannot: Design and product taste that makes the tool genuinely usable, beyond merely functional. Workflow design that bakes in the end user's expertise exactly where the model needs it. Evals built from real domain work , rather than generic benchmarks. The payoff is discipline. The leapfrog becomes repeatable where the organisation can define quality, measure it, and keep expert judgment in the path of the work.\n\nWhat custom adds\n01 Company context enters first Knowledge, examples, policies, customer language, and edge cases are part of the workflow before generation starts. 02 Expert review is designed in The system asks for judgment at the points where judgment changes the answer, before damage is done. 03 Quality becomes measurable Evaluation sets, audit trails, and acceptance criteria turn taste from a private preference into an operating standard. 04 The workflow ships AI moves from a generic chat box to a domain-shaped application that can be improved, governed, and reused. Fig 13 / Custom applications put expert judgment, company context, and measurement inside the workflow. 15 The hope\n\nAI as an amplifier of taste.\nModels will keep improving, but judgment stays human. People with talent who learn to direct the model will make work they could not have made alone. The future worth building is one where AI raises the ceiling for the people who care about the work. Map where judgment drops out Return to the operator map The Taste Gap", + "links": [], + "chunks": [ + { + "chunk_id": "tai-88dbe4d251241bb3f56948e8", + "text": "The Taste Gap 1 The question\n\nAI is everywhere. AI work that lands is rare. Why?\nThe gap between AI slop and AI work has become a problem of execution, not capability. AI can be used to increase the quality of your work, but for many it is reducing it. To produce a great output, and not just a passable one, AI needs your help. The teams pulling ahead are putting expert judgment back into the workflow at every step. Some of that work requires structured AI training. Some happens in building custom AI systems. Towards AI Research / 2026 2 The map\n\nTwo questions, four kinds of operator.\nTwo questions split the field. Does the operator have taste at the task? Do they bring that taste into the AI workflow? The upper-right box is where the leapfrog is most likely. The expert keeps judgment in the loop, rejects the first draft, and turns the model into a tool for range, speed, and finish. Most AI debate confuses these boxes, and most companies do too. Many rollouts accidentally move experts into the trap: the model gets the task, but the expert's judgment never reaches the work. HAS TASTE AT TASK →\n\nHas taste, skips it\nThe expert who outsources judgment to the model. Output drifts back toward average. THE TRAP\n\nHas taste, uses it\nDirection at every step. Work gains range, speed, and finish without losing judgment. THE WIN\n\nNo taste, no AI\nBelow a useful baseline and stays there. The pre-AI default for non-experts at this task. UNCHANGED\n\nNo taste, uses AI\nLifted to average. A real and useful gain. This is most users today. THE LIFT INJECTS TASTE INTO AI WORKFLOW → Fig 02 / The operating model behind good and bad AI work. 3 The baseline", + "heading": "Two questions, four kinds of operator.", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-89cf46d10cf920c0b155cb0f", + "text": "The Taste Gap" + }, + { + "span_id": "span-033dd9fcca2ae2bff0870d0a", + "text": "1 The question" + }, + { + "span_id": "span-2c895da9a67f0704376dac18", + "text": "The gap between AI slop and AI work has become a problem of execution, not capability." + }, + { + "span_id": "span-e02bb656b74cd6993c3bda76", + "text": "AI can be used to increase the quality of your work, but for many it is reducing it." + }, + { + "span_id": "span-70c3e7ace0d7b193d9403cd1", + "text": "To produce a great output, and not just a passable one, AI needs your help." + }, + { + "span_id": "span-23f61fbdec6fae42b42ea987", + "text": "The teams pulling ahead are putting expert judgment back into the workflow at every step." + }, + { + "span_id": "span-ed578de10dd9d16761dc888b", + "text": "Some of that work requires structured AI training." + }, + { + "span_id": "span-6b6ec5514676a205c57a17f2", + "text": "Some happens in building custom AI systems." + }, + { + "span_id": "span-e1b04ed835ac1459dbe744d8", + "text": "Towards AI Research / 2026" + }, + { + "span_id": "span-beec36fa2ea71b98b493a8bc", + "text": "2 The map" + }, + { + "span_id": "span-28f028ea3b8d3208120dd8dc", + "text": "Two questions split the field." + }, + { + "span_id": "span-849a73b6a92e8a1f2655a37c", + "text": "Does the operator have taste at the task?" + }, + { + "span_id": "span-b842cb3f0b8db4423641acbc", + "text": "Do they bring that taste into the AI workflow?" + }, + { + "span_id": "span-fd9300a0ed8012b10f934d9c", + "text": "The upper-right box is where the leapfrog is most likely." + }, + { + "span_id": "span-56b6ead9b64ab204dd9fbc2e", + "text": "The expert keeps judgment in the loop, rejects the first draft, and turns the model into a tool for range, speed, and finish." + }, + { + "span_id": "span-7321a7cd8c5c8d5bf775e5e0", + "text": "Most AI debate confuses these boxes, and most companies do too." + }, + { + "span_id": "span-53544a3c70b576da6dce6f91", + "text": "Many rollouts accidentally move experts into the trap: the model gets the task, but the expert's judgment never reaches the work." + }, + { + "span_id": "span-19f09cd293de07c1cfbc5622", + "text": "HAS TASTE AT TASK →" + }, + { + "span_id": "span-888ab3862d91481eaccfba34", + "text": "The expert who outsources judgment to the model." + }, + { + "span_id": "span-ce889772ce963db3c950117d", + "text": "Output drifts back toward average." + }, + { + "span_id": "span-c960921aff15516fb22b80ed", + "text": "THE TRAP" + }, + { + "span_id": "span-64550b324af4835ac6520bf1", + "text": "Direction at every step." + }, + { + "span_id": "span-14f4f69fabe746d016c5c78a", + "text": "Work gains range, speed, and finish without losing judgment." + }, + { + "span_id": "span-a619a49599f8b6236e5900ff", + "text": "THE WIN" + }, + { + "span_id": "span-81bed76d1dd3331fb5f6f770", + "text": "Below a useful baseline and stays there." + }, + { + "span_id": "span-9d98dfaeb023bbd2487a6cdd", + "text": "The pre-AI default for non-experts at this task." + }, + { + "span_id": "span-a4a2b994f198549fd15ecbdb", + "text": "Lifted to average." + }, + { + "span_id": "span-e7dce8ffb06e0ddbe6fc5f7f", + "text": "A real and useful gain." + }, + { + "span_id": "span-1f5c08176364cdb81050dde3", + "text": "This is most users today." + }, + { + "span_id": "span-f736ae8ab2b882cf44c9a03e", + "text": "THE LIFT" + }, + { + "span_id": "span-5ee28d2bca6d4e7aa4ebc94a", + "text": "INJECTS TASTE INTO AI WORKFLOW →" + }, + { + "span_id": "span-c941d461a625d7a3351a202a", + "text": "Fig 02 / The operating model behind good and bad AI work." + }, + { + "span_id": "span-cbb0eb9d9c711274dc3ca930", + "text": "3 The baseline" + } + ] + }, + { + "chunk_id": "tai-1a9da8492cdb45ac7bd37c86", + "text": "Generic prompts tend toward generic output.\nThe model was trained on a large body of text, pictures, and code. Without direction, it tends to return the patterns it has seen most often. It scores many possible next words and turns those scores into probabilities. In unguided use, familiar openings, safe structures, and obvious punchlines often sit near the top. That default behaviour matters. Interesting work usually begins when a person changes the path with taste, constraints, examples, rejected drafts, and sharper standards. Run the prompt cold and you get the median answer, polished into clean sentences. Fig 03 / The bell curve shows how skill is distributed. The AI baseline is the vertical line at the mean. 4 The lift\n\nFor most people, average is a promotion.\nAt many specific tasks, people sit below a useful baseline. A support agent writes a clearer refund note. A founder turns messy notes into a tolerable investor update. A junior analyst gets the first SQL query working instead of staring at a blank editor. This is real value. The model can supply structure, grammar, syntax, and a first pass at the thing the person could not yet make alone. This is the easiest part of the AI story to defend, and the part most often dismissed. Baseline competence matters, even when the result is useful rather than distinctive. Fig 04 / Each dot is one person. Below-baseline people rise vertically to the average. 5 The flood", + "heading": "Generic prompts tend toward generic output.", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-e2d8e5c66b2d5a7ad469f5fa", + "text": "The model was trained on a large body of text, pictures, and code." + }, + { + "span_id": "span-4b613938974c0c7740d05ef0", + "text": "Without direction, it tends to return the patterns it has seen most often." + }, + { + "span_id": "span-05beef8e93d2bea151ee658d", + "text": "It scores many possible next words and turns those scores into probabilities." + }, + { + "span_id": "span-d59d93d2a44e956897138be7", + "text": "In unguided use, familiar openings, safe structures, and obvious punchlines often sit near the top." + }, + { + "span_id": "span-9a49d4448fc9110b8cc18924", + "text": "That default behaviour matters." + }, + { + "span_id": "span-cde9256444124eb4fe89bb51", + "text": "Interesting work usually begins when a person changes the path with taste, constraints, examples, rejected drafts, and sharper standards." + }, + { + "span_id": "span-156fe3ef279cc4880a3f453d", + "text": "Run the prompt cold and you get the median answer, polished into clean sentences." + }, + { + "span_id": "span-f9aece06037112b40740fd1c", + "text": "Fig 03 / The bell curve shows how skill is distributed." + }, + { + "span_id": "span-506c2033c546a92cdba67b47", + "text": "The AI baseline is the vertical line at the mean." + }, + { + "span_id": "span-1fc5d7477ae3697bc0bb4dd4", + "text": "4 The lift" + }, + { + "span_id": "span-169b96debd0439b7c8a664e7", + "text": "At many specific tasks, people sit below a useful baseline." + }, + { + "span_id": "span-c9037e9575846beed7616aca", + "text": "A support agent writes a clearer refund note." + }, + { + "span_id": "span-7d81a1ae7fbb27f0afa91845", + "text": "A founder turns messy notes into a tolerable investor update." + }, + { + "span_id": "span-0037056d8b7cf914cec903a4", + "text": "A junior analyst gets the first SQL query working instead of staring at a blank editor." + }, + { + "span_id": "span-6861068daa8de601eecdfd3f", + "text": "This is real value." + }, + { + "span_id": "span-a90a699f95998bb6f677c1b4", + "text": "The model can supply structure, grammar, syntax, and a first pass at the thing the person could not yet make alone." + }, + { + "span_id": "span-9ebc519bb5e5598a9ef509e4", + "text": "This is the easiest part of the AI story to defend, and the part most often dismissed." + }, + { + "span_id": "span-2ddf8c7cfb31adb703a5c401", + "text": "Baseline competence matters, even when the result is useful rather than distinctive." + }, + { + "span_id": "span-c396479aa0cfdef8283aa8b4", + "text": "Fig 04 / Each dot is one person." + }, + { + "span_id": "span-3ce9b20477c8f17464ef49df", + "text": "Below-baseline people rise vertically to the average." + }, + { + "span_id": "span-7c4259a0ea7ee7925f202a76", + "text": "5 The flood" + } + ] + }, + { + "chunk_id": "tai-27f2482aeefad3423a8e89fe", + "text": "AI slop is repetition at scale.\nA writing quirk can be charming when one person owns it. A familiar cadence, tidy transition, stock joke, or neat three-part structure may pass unnoticed in a single piece. The irritation starts when a million people ship the same move. The phrase stops reading like style and starts reading like residue. When sameness is everywhere, average output becomes brand drag, review burden, and operational noise. Fig 05 / One repeated move can be harmless. At scale, the pattern becomes the problem. 6 The trap\n\nThe expert who hands the keys over.\nAn expert has real taste built over years. They open the model, type a generic prompt, and ship whatever comes back. The output may look polished, but the expert is no longer present in the work. Their edge came from judgment. The prompt removed it. What used to be their value is now a setting they did not change. Skill that does not enter the workflow does not enter the work. Fig 06 / The expert who skips the taste step ends up exactly where everyone else ends up. 7 The leapfrog", + "heading": "AI slop is repetition at scale.", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-f7f59643b71ac194103a1b7e", + "text": "A writing quirk can be charming when one person owns it." + }, + { + "span_id": "span-8b9370c7576087f977490d7a", + "text": "A familiar cadence, tidy transition, stock joke, or neat three-part structure may pass unnoticed in a single piece." + }, + { + "span_id": "span-821fa03a58739889d1cb289f", + "text": "The irritation starts when a million people ship the same move." + }, + { + "span_id": "span-f6801d43c385624526b7c4aa", + "text": "The phrase stops reading like style and starts reading like residue." + }, + { + "span_id": "span-2e5485b72577d1b29a82026b", + "text": "When sameness is everywhere, average output becomes brand drag, review burden, and operational noise." + }, + { + "span_id": "span-f4bcf102107c324d077f6b03", + "text": "Fig 05 / One repeated move can be harmless." + }, + { + "span_id": "span-cf1dabe60dacdb3e15944d9b", + "text": "At scale, the pattern becomes the problem." + }, + { + "span_id": "span-1b60a863ba6c89d6ebda07e4", + "text": "6 The trap" + }, + { + "span_id": "span-d3110dafab45d8f1fb04442a", + "text": "An expert has real taste built over years." + }, + { + "span_id": "span-d5b0e5d9a1497ae6b7b1eeec", + "text": "They open the model, type a generic prompt, and ship whatever comes back." + }, + { + "span_id": "span-3a9a6b1c2c1eb1273a1b9412", + "text": "The output may look polished, but the expert is no longer present in the work." + }, + { + "span_id": "span-816b83309e50da896d6dd5a1", + "text": "Their edge came from judgment." + }, + { + "span_id": "span-a45efa065cfb2fa70820f048", + "text": "The prompt removed it." + }, + { + "span_id": "span-cd89c627976aec9697314175", + "text": "What used to be their value is now a setting they did not change." + }, + { + "span_id": "span-880053c5ec69442937221383", + "text": "Skill that does not enter the workflow does not enter the work." + }, + { + "span_id": "span-50ca4cdd6dcbd6ec5a8c53b4", + "text": "Fig 06 / The expert who skips the taste step ends up exactly where everyone else ends up." + }, + { + "span_id": "span-ccf4cae8930609cc59f422cd", + "text": "7 The leapfrog" + } + ] + }, + { + "chunk_id": "tai-8feca65b05e3b102b90c7eef", + "text": "The expert who directs the loop.\nA smaller group does the harder thing. They bring decades of career experience: real customer conversations, hard-won edge cases, and industry memory that was never cleanly captured in the model's training set. Then they reject the first draft, ask for alternatives, compare them, name what is wrong, and drive it into a second shape. AI has jagged intelligence: superhuman in some narrow moves, weak in places a good operator spots immediately. The leapfrog happens when the model supplies range and speed while the human supplies taste, context, touch, and judgment. The reader cannot always name why, but the work feels chosen rather than produced. Fig 07 / The arrow visibly arcs over the holdout. Same starting skill, very different finish. 8 The vibe gap", + "heading": "The expert who directs the loop.", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-2673d84bad2ac6fcc9c62285", + "text": "A smaller group does the harder thing." + }, + { + "span_id": "span-994873941f6cb56a40c9005a", + "text": "They bring decades of career experience: real customer conversations, hard-won edge cases, and industry memory that was never cleanly captured in the model's training set." + }, + { + "span_id": "span-bad3164862cb37c5a7e0eb09", + "text": "Then they reject the first draft, ask for alternatives, compare them, name what is wrong, and drive it into a second shape." + }, + { + "span_id": "span-881952055bb4d88c2e19fa12", + "text": "AI has jagged intelligence: superhuman in some narrow moves, weak in places a good operator spots immediately." + }, + { + "span_id": "span-70333f5982af18ae842dbc47", + "text": "The leapfrog happens when the model supplies range and speed while the human supplies taste, context, touch, and judgment." + }, + { + "span_id": "span-19f2213e2b3b3fc9aab870ca", + "text": "The reader cannot always name why, but the work feels chosen rather than produced." + }, + { + "span_id": "span-8c7a8471d0cfba530765df06", + "text": "Fig 07 / The arrow visibly arcs over the holdout." + }, + { + "span_id": "span-de6c4259c3b2b1181057b015", + "text": "Same starting skill, very different finish." + }, + { + "span_id": "span-0e1ccd0a80eb212e1c951937", + "text": "8 The vibe gap" + } + ] + }, + { + "chunk_id": "tai-5b1bd62d041b292656586745", + "text": "Vibe coding makes everyone look like a builder.\nA non-engineer can vibe-code a working app in an afternoon. A non-designer can produce a credible landing page in an hour. A non-writer can put a long report on the page by lunch. The speed gains are real and unevenly distributed: people who used to be locked out of building the artifact at all are now shipping demos. For first-draft work, AI compressed the visible part of the job by something like an order of magnitude, and that compression is not going away. The artifact is not the deliverable. A demoable app still has to choose its state model, handle auth and rate limits, survive a flaky network, deal with the empty state and the error state, log enough to debug an outage, and make the small interface decisions that decide whether anyone returns next week. A polished report still needs the source check, the contradicting study the model did not surface, the structural cut, the falsifiable claim, and the line that earns the close. The discount on that work is much smaller than the discount on the first draft. That second layer still belongs to expert hours. The total project is faster than it used to be, but the savings curve is steepest at the start and flattens hard once the surface looks done. A vibe-coded prototype that took three hours can take another forty before it scales as a product. A vibe-written report that took ninety minutes can take another two days before it survives review. The ratio inverts: most of the remaining time sits in the part nobody outside the team can see. That is where the manager trap lives. The person who built the prototype reads it as ninety percent done. The engineer or editor who can name what is missing reads it as forty, and the schedule is being set by the higher number. The same", + "heading": "Vibe coding makes everyone look like a builder.", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-beb5fd4d70695e138d0a5cce", + "text": "A non-engineer can vibe-code a working app in an afternoon." + }, + { + "span_id": "span-b87e7bd30418b567befb31b3", + "text": "A non-designer can produce a credible landing page in an hour." + }, + { + "span_id": "span-cb44989e0f3278c9f09af332", + "text": "A non-writer can put a long report on the page by lunch." + }, + { + "span_id": "span-9d1187a6c0fbcf0021be6503", + "text": "The speed gains are real and unevenly distributed: people who used to be locked out of building the artifact at all are now shipping demos." + }, + { + "span_id": "span-39545fd7d8514dfee9cb1daa", + "text": "For first-draft work, AI compressed the visible part of the job by something like an order of magnitude, and that compression is not going away." + }, + { + "span_id": "span-611d64380ccfac449ed283f8", + "text": "The artifact is not the deliverable." + }, + { + "span_id": "span-78aa0146ee761ffeaee3d4c2", + "text": "A demoable app still has to choose its state model, handle auth and rate limits, survive a flaky network, deal with the empty state and the error state, log enough to debug an outage, and make the small interface decisions that decide whether anyone returns next week." + }, + { + "span_id": "span-7574c8adb75d4815386b4fb9", + "text": "A polished report still needs the source check, the contradicting study the model did not surface, the structural cut, the falsifiable claim, and the line that earns the close." + }, + { + "span_id": "span-644d68f031aca1e02d1fa01c", + "text": "The discount on that work is much smaller than the discount on the first draft." + }, + { + "span_id": "span-051755b4647073751a91b426", + "text": "That second layer still belongs to expert hours." + }, + { + "span_id": "span-8b6e2aca6aeb970a67575cb8", + "text": "The total project is faster than it used to be, but the savings curve is steepest at the start and flattens hard once the surface looks done." + }, + { + "span_id": "span-958394baaa93314051266145", + "text": "A vibe-coded prototype that took three hours can take another forty before it scales as a product." + }, + { + "span_id": "span-8463d31bf0a6334dc26b8c56", + "text": "A vibe-written report that took ninety minutes can take another two days before it survives review." + }, + { + "span_id": "span-9d82c3b23c4a0479caeb36fe", + "text": "The ratio inverts: most of the remaining time sits in the part nobody outside the team can see." + }, + { + "span_id": "span-9a5a8125ee3c54cdb1526736", + "text": "That is where the manager trap lives." + }, + { + "span_id": "span-fc8d06c18a23a2a93babb8e9", + "text": "The person who built the prototype reads it as ninety percent done." + }, + { + "span_id": "span-1f1f7036245fcf11410348a8", + "text": "The engineer or editor who can name what is missing reads it as forty, and the schedule is being set by the higher number." + } + ] + }, + { + "chunk_id": "tai-9dc92284bf7e5ed4540410a2", + "text": "Vibe coding makes everyone look like a builder.\npattern shows up in design comps that fall apart on a single user test, in research output that breaks on the first peer challenge, and in marketing copy that does not survive a legal read. Vibe coding makes everyone look like a builder. The expert is the one who knows what is still missing. 9 The wrong question\n\nThe audience asks the wrong question.\nThe flood of average AI work has bred suspicion of all AI work. Readers, viewers, hiring managers, editors. The reflex is to ask whether a tool was used at all. The reflex has a real cause. Most published AI work today comes from free, weaker models. A smaller share comes from strong models used without judgment in the loop. Either way, the median output is generic, and the suspicion is earned. The trap is letting that suspicion become a tool test rather than a quality test. Agentic coding has already broken that reflex. No serious person sneers anymore because a developer did not type every line by hand. The useful signal is whether the person understood the system, made the right tradeoffs, and owned the result. The same standard should apply outside code. Ask whether human judgment shaped the work. \"Did you use AI?\" Was there human judgment in the loop? Fig 08 / Provenance is the weaker question. Judgment is the stronger one. 10 AI detectors", + "heading": "The audience asks the wrong question.", + "index": 5, + "evidence_spans": [ + { + "span_id": "span-3e76ab0e69a00530d4433ef8", + "text": "Vibe coding makes everyone look like a builder." + }, + { + "span_id": "span-43334577b571ca36c66ad713", + "text": "The expert is the one who knows what is still missing." + }, + { + "span_id": "span-a3b041a7a94a7d66a142589e", + "text": "9 The wrong question" + }, + { + "span_id": "span-ca0ee4e2ee4be7b36fbed50e", + "text": "The flood of average AI work has bred suspicion of all AI work." + }, + { + "span_id": "span-1de2003a8a1f5ba579aba72c", + "text": "Readers, viewers, hiring managers, editors." + }, + { + "span_id": "span-06385bff771a8480182a4bea", + "text": "The reflex is to ask whether a tool was used at all." + }, + { + "span_id": "span-47cb9470d654fdf90efd56f4", + "text": "The reflex has a real cause." + }, + { + "span_id": "span-907fa6e0be5cb8236400728f", + "text": "Most published AI work today comes from free, weaker models." + }, + { + "span_id": "span-7207b793c7fcf614f948ea15", + "text": "A smaller share comes from strong models used without judgment in the loop." + }, + { + "span_id": "span-42b1bf0a5e5ffc0112c209e2", + "text": "Either way, the median output is generic, and the suspicion is earned." + }, + { + "span_id": "span-c3e679b46da07d1819a0c025", + "text": "The trap is letting that suspicion become a tool test rather than a quality test." + }, + { + "span_id": "span-b16c7130eaf4c7748836d5bb", + "text": "Agentic coding has already broken that reflex." + }, + { + "span_id": "span-74a18452b10a8601b7c8b8bf", + "text": "No serious person sneers anymore because a developer did not type every line by hand." + }, + { + "span_id": "span-685a4cca463bb05bde1b309f", + "text": "The useful signal is whether the person understood the system, made the right tradeoffs, and owned the result." + }, + { + "span_id": "span-46507d5ad3e35b4e54fa3b5c", + "text": "The same standard should apply outside code." + }, + { + "span_id": "span-8aef88d229a879025b01e875", + "text": "Ask whether human judgment shaped the work." + }, + { + "span_id": "span-525da53d9357b1897f687608", + "text": "\"Did you use AI?\"" + }, + { + "span_id": "span-149a5fde1a428e6fceb11687", + "text": "Was there human judgment in the loop?" + }, + { + "span_id": "span-286259df6b165dd19cd04806", + "text": "Fig 08 / Provenance is the weaker question." + }, + { + "span_id": "span-fe70c24fbc592ec2e64671b8", + "text": "Judgment is the stronger one." + }, + { + "span_id": "span-e99611ac35cc6df55e1ba8cc", + "text": "10 AI detectors" + } + ] + }, + { + "chunk_id": "tai-97223963a359be7f1c3bdbf6", + "text": "AI detectors solve the wrong puzzle.\nSet aside the long history of false positives. Assume AI detectors worked perfectly. They would tell you whether tokens look machine-generated. They would not tell you whether a person with judgment shaped the result. The dimension everyone cares about, quality and direction, is the one thing the tool cannot see. Editors, teachers, and hiring managers using detectors are sorting on the wrong axis. DETECTOR VERDICT AI probability 92.4% Tokens flagged 147 / 612 Conclusion LIKELY AI Not measured: Human judgment, taste, structure, originality, point of view, accuracy, purpose. Fig 09 / Confidence on the wrong axis. 11 Practical hygiene\n\nEven so, scrub the obvious tells.\nYou can put real insight and effort into an AI collaboration and still lose the reader at the surface. A few familiar AI signatures can make careful work look careless. Even the strong models leak the same signatures. The tics are model-wide, not a sign of weak models. Careful work that does not get scrubbed reads like the median and gets judged that way. If you want the work judged on its merits, strip the signals that mean nothing about the work and everything about the source.\n\nSurface signals that weaken trust\nEm dashes used as a tic Forced contrast cadence Over-polished stock vocabulary Three bullet conclusions to every section Unnecessary symmetric tricolons Hedge phrases in every paragraph Generic \"in conclusion\" wrap-ups Fig 10 / Surface fixes that buy you a fair reading. 12 The bottleneck unlock", + "heading": "AI detectors solve the wrong puzzle.", + "index": 6, + "evidence_spans": [ + { + "span_id": "span-50cf7e825eb5aa2f22674819", + "text": "Set aside the long history of false positives." + }, + { + "span_id": "span-391bf069974b6425e438ba09", + "text": "Assume AI detectors worked perfectly." + }, + { + "span_id": "span-b8ffb6bfe36f21adc18e3e8d", + "text": "They would tell you whether tokens look machine-generated." + }, + { + "span_id": "span-80bd6a2f2ff5c5905666437e", + "text": "They would not tell you whether a person with judgment shaped the result." + }, + { + "span_id": "span-9500ab19b06220f612337168", + "text": "The dimension everyone cares about, quality and direction, is the one thing the tool cannot see." + }, + { + "span_id": "span-24f4810aae078b16028e5866", + "text": "Editors, teachers, and hiring managers using detectors are sorting on the wrong axis." + }, + { + "span_id": "span-6156086e7f4bfc6f2b96046a", + "text": "DETECTOR VERDICT" + }, + { + "span_id": "span-0b399ff0ea49b1cdc54f7831", + "text": "AI probability 92.4%" + }, + { + "span_id": "span-d3a6ba4d5a7faa26449666c8", + "text": "Tokens flagged 147 / 612" + }, + { + "span_id": "span-ef4276ae10640d3e6f465970", + "text": "Conclusion LIKELY AI" + }, + { + "span_id": "span-dcd37606733670afecab5958", + "text": "Not measured: Human judgment, taste, structure, originality, point of view, accuracy, purpose." + }, + { + "span_id": "span-7241c8a5cb9a911edd9c4945", + "text": "Fig 09 / Confidence on the wrong axis." + }, + { + "span_id": "span-328a0532bb40290845049893", + "text": "11 Practical hygiene" + }, + { + "span_id": "span-f8f883a1dee389a32a25c34e", + "text": "You can put real insight and effort into an AI collaboration and still lose the reader at the surface." + }, + { + "span_id": "span-ae04a9c950b770dd7042e8f9", + "text": "A few familiar AI signatures can make careful work look careless." + }, + { + "span_id": "span-b4681619009e952c783c8bdb", + "text": "Even the strong models leak the same signatures." + }, + { + "span_id": "span-a13edf7e648f4f51ccae6d78", + "text": "The tics are model-wide, not a sign of weak models." + }, + { + "span_id": "span-e0ad7597014921962aba6625", + "text": "Careful work that does not get scrubbed reads like the median and gets judged that way." + }, + { + "span_id": "span-cff8bfd8db15db85456f6710", + "text": "If you want the work judged on its merits, strip the signals that mean nothing about the work and everything about the source." + }, + { + "span_id": "span-3522b0d0d7ddece703c79252", + "text": "Em dashes used as a tic" + }, + { + "span_id": "span-7e171faad2a61d03e12400cc", + "text": "Forced contrast cadence" + }, + { + "span_id": "span-67e6af7261749afa9eb34d60", + "text": "Over-polished stock vocabulary" + }, + { + "span_id": "span-2bd0647ee8e4683f1813ea31", + "text": "Three bullet conclusions to every section" + }, + { + "span_id": "span-f3b1bcc2b81239578b7a731f", + "text": "Unnecessary symmetric tricolons" + }, + { + "span_id": "span-d34e07692255c9c16f3850bd", + "text": "Hedge phrases in every paragraph" + }, + { + "span_id": "span-3e64ce5055e7992cb181eae7", + "text": "Generic \"in conclusion\" wrap-ups" + }, + { + "span_id": "span-d4e185befc8612a98ee534f2", + "text": "Fig 10 / Surface fixes that buy you a fair reading." + }, + { + "span_id": "span-cbe096ffb15b39a59aded7a0", + "text": "12 The bottleneck unlock" + } + ] + }, + { + "chunk_id": "tai-29081ff9d19d8fb37ccf95aa", + "text": "World-class hidden human talent can be revealed by AI.\nA great output requires a package of skills; many people have top-tier talent in one area that never reaches an audience because their surrounding skills are subpar. The weakest skill sets the ceiling. A person can be world-class at one thing and remain blocked because the surrounding skills sit below the publishable line. AI can lift the supporting skills above that line, and the talent finally reaches an audience. Rare talent Clears bar Below bar Lifted by AI Songwriter · generational hooks Bottleneck: vocals + production Writes melodies people will hum for years. Cannot sing them. Has no idea how to mix. Without AI Does not ship Melody Lyrics ✕ Vocals ✕ Mix Two failing skills keep the songs in a notebook. With AI Ships full package Melody Lyrics Vocals Mix AI vocals and mix. The hooks finally reach listeners. Operations lead · rare process judgment Bottleneck: tooling + reporting Knows where work gets stuck. Cannot turn that knowledge into a reliable internal tool. Without AI Manual process Process ✕ Data ✕ Tooling ✕ Reports Expert time leaks into cleanup, reminders, and status chasing. With AI Workflow ships Process Data Tooling Reports A custom workflow fills the gaps. Process judgment becomes operational leverage. Fig 11 / The dashed line is the publishable bar. The weakest leg decides if the package ships. 13 The maturity curve", + "heading": "World-class hidden human talent can be revealed by AI.", + "index": 7, + "evidence_spans": [ + { + "span_id": "span-73e03474e328326e6ba6325e", + "text": "A great output requires a package of skills; many people have top-tier talent in one area that never reaches an audience because their surrounding skills are subpar." + }, + { + "span_id": "span-935f8ce64047fa025994dc3d", + "text": "The weakest skill sets the ceiling." + }, + { + "span_id": "span-eef24dfb4f6d880daec54c71", + "text": "A person can be world-class at one thing and remain blocked because the surrounding skills sit below the publishable line." + }, + { + "span_id": "span-c976ab919ee46880c2c7cb71", + "text": "AI can lift the supporting skills above that line, and the talent finally reaches an audience." + }, + { + "span_id": "span-988b71b5b232344ac882045d", + "text": "Rare talent Clears bar Below bar Lifted by AI" + }, + { + "span_id": "span-c4fa317f79a6d5cc15601a9f", + "text": "Songwriter · generational hooks" + }, + { + "span_id": "span-1089d5918e87ff88dfcc84ac", + "text": "Bottleneck: vocals + production" + }, + { + "span_id": "span-e31a713fc7ff2a720465dc34", + "text": "Writes melodies people will hum for years." + }, + { + "span_id": "span-c094b8498615ca8f30e4a74b", + "text": "Cannot sing them." + }, + { + "span_id": "span-46560ce7c14acabc06f540ae", + "text": "Has no idea how to mix." + }, + { + "span_id": "span-ad95cb56ba4d42607b387f15", + "text": "Without AI Does not ship" + }, + { + "span_id": "span-a7e9ec3c025420ec8c26aa68", + "text": "Two failing skills keep the songs in a notebook." + }, + { + "span_id": "span-87f083341efe29051321a40c", + "text": "With AI Ships full package" + }, + { + "span_id": "span-ab164b777454b52ae831b0b5", + "text": "AI vocals and mix." + }, + { + "span_id": "span-de6ab73e3e887dad5ebd8fba", + "text": "The hooks finally reach listeners." + }, + { + "span_id": "span-49a2f2f94a7233d968fdaccf", + "text": "Operations lead · rare process judgment" + }, + { + "span_id": "span-19644d3adb290f3f36891fb2", + "text": "Bottleneck: tooling + reporting" + }, + { + "span_id": "span-83cc3f6cb5ce4e41be13b5d9", + "text": "Knows where work gets stuck." + }, + { + "span_id": "span-2d327de2f9b333d1cd7d6477", + "text": "Cannot turn that knowledge into a reliable internal tool." + }, + { + "span_id": "span-4bf4806009a72879434d08f0", + "text": "Without AI Manual process" + }, + { + "span_id": "span-03c760e6bd1c97326265d377", + "text": "Expert time leaks into cleanup, reminders, and status chasing." + }, + { + "span_id": "span-bff3800c25fd535773b04123", + "text": "With AI Workflow ships" + }, + { + "span_id": "span-a9c1a3c055bf830cd92c525d", + "text": "A custom workflow fills the gaps." + }, + { + "span_id": "span-8a0374205072d34007c87e99", + "text": "Process judgment becomes operational leverage." + }, + { + "span_id": "span-a49c50220e2a5884326d97ae", + "text": "Fig 11 / The dashed line is the publishable bar." + }, + { + "span_id": "span-adb3088d2070becfe393a18a", + "text": "The weakest leg decides if the package ships." + }, + { + "span_id": "span-a3a0c82f77a6a923b922592d", + "text": "13 The maturity curve" + } + ] + }, + { + "chunk_id": "tai-f18951c7e63bc287e5b34740", + "text": "Companies move through four stages of AI adoption.\nStage zero is familiar work without AI. It is slower, but the accountability chain is clear. People know who made the call and where judgment entered the work. Stage one is shadow AI. People grab free tools, paste work into generic interfaces, and invent private workflows. Quality can dip below the pre-AI line when experts delegate judgment instead of applying it. Risk rises too, from data leakage to brand damage to compliance gaps. Stage two is managed AI. Sanctioned tools, paid licences, policies, and training pull quality back up. The company is using AI on purpose, but much of the work still happens inside generic interfaces. Stage three is embedded AI. Custom workflows are built into core processes, with expert review points, domain rules, retrieval, evals, and audit trails, and the leapfrog from earlier sections is scaled across an organisation. The goal is higher-quality output from humans and AI together. Speed comes for free at every stage. Quality only rises when human judgment lives inside the system rather than arriving as cleanup. Fig 12 / Shadow AI is where risk rises before quality systems catch up. 14 From general to custom", + "heading": "Companies move through four stages of AI adoption.", + "index": 8, + "evidence_spans": [ + { + "span_id": "span-a65d17f3f6403fb38dbc191b", + "text": "Stage zero is familiar work without AI." + }, + { + "span_id": "span-95392257ca331b14a6fa269e", + "text": "It is slower, but the accountability chain is clear." + }, + { + "span_id": "span-58011201fb406a3afd36c119", + "text": "People know who made the call and where judgment entered the work." + }, + { + "span_id": "span-52ada6adaa6400383fad423b", + "text": "Stage one is shadow AI." + }, + { + "span_id": "span-88cbbec22f2a3d6c70a12e55", + "text": "People grab free tools, paste work into generic interfaces, and invent private workflows." + }, + { + "span_id": "span-9c226560a15d22df196452e9", + "text": "Quality can dip below the pre-AI line when experts delegate judgment instead of applying it." + }, + { + "span_id": "span-87d6a02443c1c27ade7cf4a3", + "text": "Risk rises too, from data leakage to brand damage to compliance gaps." + }, + { + "span_id": "span-8c3eb3e209ca524bb0006b57", + "text": "Stage two is managed AI." + }, + { + "span_id": "span-549dae4d675e63aa5c501d6e", + "text": "Sanctioned tools, paid licences, policies, and training pull quality back up." + }, + { + "span_id": "span-371b207a8d2cc830a41c3ae4", + "text": "The company is using AI on purpose, but much of the work still happens inside generic interfaces." + }, + { + "span_id": "span-8156691574553a6f7f1d530a", + "text": "Stage three is embedded AI." + }, + { + "span_id": "span-3971df78a19c1c702d85a7f6", + "text": "Custom workflows are built into core processes, with expert review points, domain rules, retrieval, evals, and audit trails, and the leapfrog from earlier sections is scaled across an organisation." + }, + { + "span_id": "span-8c65576daf9f607e0bb70d85", + "text": "The goal is higher-quality output from humans and AI together." + }, + { + "span_id": "span-00320339996dfb1501a0654c", + "text": "Speed comes for free at every stage." + }, + { + "span_id": "span-981bed0cb570de3a70cce24d", + "text": "Quality only rises when human judgment lives inside the system rather than arriving as cleanup." + }, + { + "span_id": "span-7d782bd7967fe34a0ac3a102", + "text": "Fig 12 / Shadow AI is where risk rises before quality systems catch up." + }, + { + "span_id": "span-233df3c71c2236d39bc0420b", + "text": "14 From general to custom" + } + ] + }, + { + "chunk_id": "tai-dbbf18beef5c0e8a9f07b7ad", + "text": "The next jump is from general tools to custom applications.\nEnterprise adoption is real, but many companies still use the same generic interface for specialized work. That leaves expert judgment outside the system, where it arrives late as correction, cleanup, or rejection. A generic chatbot can answer a question. A vibe-coded app can demo an idea. Neither automatically carries an engineer's taste for states, failure modes, latency, permissions, data shape, review loops, or the small interface decisions that make a solution feel inevitable. The larger value sits in applications built around the work itself: role-specific interfaces, company knowledge, domain rules, human review points, evals, audit trails, and integrations into existing systems. What an engineering team adds that a generic chatbox or quick demo cannot: Design and product taste that makes the tool genuinely usable, beyond merely functional. Workflow design that bakes in the end user's expertise exactly where the model needs it. Evals built from real domain work , rather than generic benchmarks. The payoff is discipline. The leapfrog becomes repeatable where the organisation can define quality, measure it, and keep expert judgment in the path of the work.", + "heading": "The next jump is from general tools to custom applications.", + "index": 9, + "evidence_spans": [ + { + "span_id": "span-43f69f989f7d7957033a9b12", + "text": "Enterprise adoption is real, but many companies still use the same generic interface for specialized work." + }, + { + "span_id": "span-14a1e8ab351534275658c9c2", + "text": "That leaves expert judgment outside the system, where it arrives late as correction, cleanup, or rejection." + }, + { + "span_id": "span-688832e600c8ae58e9c9baf9", + "text": "A generic chatbot can answer a question." + }, + { + "span_id": "span-4303a5e40f97eca0c76d6fc8", + "text": "A vibe-coded app can demo an idea." + }, + { + "span_id": "span-2ef4a5db7620f839866de456", + "text": "Neither automatically carries an engineer's taste for states, failure modes, latency, permissions, data shape, review loops, or the small interface decisions that make a solution feel inevitable." + }, + { + "span_id": "span-0aa014448259e16b76d1e065", + "text": "The larger value sits in applications built around the work itself: role-specific interfaces, company knowledge, domain rules, human review points, evals, audit trails, and integrations into existing systems." + }, + { + "span_id": "span-ce14b1e5a755dff660e0f1f6", + "text": "What an engineering team adds that a generic chatbox or quick demo cannot:" + }, + { + "span_id": "span-4a1fb5cc69d2c38cf508905f", + "text": "Design and product taste that makes the tool genuinely usable, beyond merely functional." + }, + { + "span_id": "span-9887f2e81d75fcd12dc39995", + "text": "Workflow design that bakes in the end user's expertise exactly where the model needs it." + }, + { + "span_id": "span-b94abd78789cdfc9c211b80c", + "text": "Evals built from real domain work , rather than generic benchmarks." + }, + { + "span_id": "span-2f2fb981324b19ac706fc0c7", + "text": "The payoff is discipline." + }, + { + "span_id": "span-f2eea78237faeeaa0100efb5", + "text": "The leapfrog becomes repeatable where the organisation can define quality, measure it, and keep expert judgment in the path of the work." + } + ] + }, + { + "chunk_id": "tai-73eb57f6864ccd5a1a6070b6", + "text": "What custom adds\n01 Company context enters first Knowledge, examples, policies, customer language, and edge cases are part of the workflow before generation starts. 02 Expert review is designed in The system asks for judgment at the points where judgment changes the answer, before damage is done. 03 Quality becomes measurable Evaluation sets, audit trails, and acceptance criteria turn taste from a private preference into an operating standard. 04 The workflow ships AI moves from a generic chat box to a domain-shaped application that can be improved, governed, and reused. Fig 13 / Custom applications put expert judgment, company context, and measurement inside the workflow. 15 The hope\n\nAI as an amplifier of taste.\nModels will keep improving, but judgment stays human. People with talent who learn to direct the model will make work they could not have made alone. The future worth building is one where AI raises the ceiling for the people who care about the work. Map where judgment drops out Return to the operator map The Taste Gap", + "heading": "What custom adds", + "index": 10, + "evidence_spans": [ + { + "span_id": "span-36636276fdea0c0951741ae3", + "text": "Company context enters first Knowledge, examples, policies, customer language, and edge cases are part of the workflow before generation starts." + }, + { + "span_id": "span-76f4c66a45b07a2d1417baa7", + "text": "Expert review is designed in The system asks for judgment at the points where judgment changes the answer, before damage is done." + }, + { + "span_id": "span-dff3e305b48dabac6b26ddff", + "text": "Quality becomes measurable Evaluation sets, audit trails, and acceptance criteria turn taste from a private preference into an operating standard." + }, + { + "span_id": "span-3bd7049b357d448fbabbc7d4", + "text": "The workflow ships AI moves from a generic chat box to a domain-shaped application that can be improved, governed, and reused." + }, + { + "span_id": "span-343c5049eb174c464989d017", + "text": "Fig 13 / Custom applications put expert judgment, company context, and measurement inside the workflow." + }, + { + "span_id": "span-322a49b72c5917bbe38e8cd7", + "text": "15 The hope" + }, + { + "span_id": "span-257aeee71b0070b02e1ec930", + "text": "Models will keep improving, but judgment stays human." + }, + { + "span_id": "span-209203809320487d97cd5a4e", + "text": "People with talent who learn to direct the model will make work they could not have made alone." + }, + { + "span_id": "span-c1fe4c00bfeb319b2f134f22", + "text": "The future worth building is one where AI raises the ceiling for the people who care about the work." + }, + { + "span_id": "span-4d85243e92517e9dc3d31a3c", + "text": "Map where judgment drops out Return to the operator map" + }, + { + "span_id": "span-81f7d4e3815c6b6c8633202f", + "text": "The Taste Gap" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "e1d0346eda71f7781cc15279ac4fca8870c98a76fe99cc4ea5e2989e63c261a8", + "content_hash": "e1d0346eda71f7781cc15279ac4fca8870c98a76fe99cc4ea5e2989e63c261a8", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/theaitastegap/" + ], + "evidence_hash": "76fc95eb7cd5bfd47f25a9aba8d053c905f0a2792d8dce8209f764578060609b" + }, + { + "discovered_url": "https://towardsai.com/webinars/agentengineering/", + "url": "https://towardsai.com/webinars/agentengineering/", + "canonical_url": "https://towardsai.com/webinars/agentengineering/", + "host": "towardsai.com", + "path": "/webinars/agentengineering", + "kind": "free_resource", + "offer_id": null, + "entity_id": "page:towardsai.com/webinars/agentengineering", + "title": "Agentic AI Webinar | Become an AI Engineer | Towards AI", + "review_title": "Agent Engineering webinar", + "reviewed_summary": "Free agent engineering webinar.", + "meta_description": "Watch the recorded webinar on the architecture mistakes AI engineers make and the decision frameworks to avoid them. Hosted by Louis-Francois Bouchard, Lead AI Engineer & Co-Founder at Towards AI.", + "headings": [ + "The Architecture Mistakes AI Engineers Make", + "Are you making these AI system design mistakes?", + "Over-Engineered AI Systems", + "Discovering Requirements Late", + "Reliability Risk", + "What You'll Learn", + "A Process To Make Clean Design Decisions", + "Start With Clear Requirements", + "Design For Reliability From Day 1", + "Architect For Millions, NOT Your First Demo", + "Hi, I'm Louis-François", + "Stop Making AI Projects That Never Make it Past the Demo" + ], + "text": "Learn how to make architectural decisions\n\nThe Architecture Mistakes AI Engineers Make\n...and the decision frameworks to prevent them, trusted by engineers at Intel, Europol, Activeloop and more. Get Instant Access arrow_forward Trusted by 100,000+ engineers: ⭐⭐⭐⭐⭐ Louis-François Bouchard Lead AI Engineer & Co-Founder at Towards AI Industry Endorsement \"Helps learners understand everything from fundamentals to advanced building blocks of LLM applications. Most comprehensive resource I've seen.\" Jerry Liu CEO of LlamaIndex Engineers from global leaders are learning with us\n\nAre you making these AI system design mistakes?\nMost AI projects fail in production. Not from lack of skill, but from predictable architectural choices. 01\n\nOver-Engineered AI Systems\nDefaulting to \"agent + tools\" leads to unnecessary complexity, compounding failure points and debugging nightmares. 02\n\nDiscovering Requirements Late\nUnclear requirements cause cost blowups, tool sprawl, and untestable autonomy. 03\n\nReliability Risk\nAI systems fail in production through hallucinated outputs, edge case failures, and poor observability.\n\nWhat You'll Learn\n01\n\nA Process To Make Clean Design Decisions\nInstead of defaulting to agents + tools, you'll have a clear framework to make architectural decisions using constraints. 02\n\nStart With Clear Requirements\nLearn the kickoff questions that force clarity upfront, so cost blowups, tool sprawl, and untestable autonomy don't show up after you've already built half the system. 03\n\nDesign For Reliability From Day 1\nPractical rules for making agents production-safe dialing in a thin agent layer with validation loops, observability, and human-in-the-loop design. 04\n\nArchitect For Millions, NOT Your First Demo\nLearn to build the architecture for real user traffic, keeping your agents thin so the system is debuggable and observable in production. Get Instant Access\n\nHi, I'm Louis-François\nLead AI Engineer and Educator\n\nHi, I'm Louis-François\nLead AI Engineer and Educator I’m the co-founder of Towards AI and author of the bestselling book, Building LLMs for Production . After years of research, I dropped out of my Ph.D. because building AI teams in the startup world showed me a harsh reality: traditional universities are failing to prepare engineers for the industry. We created Towards AI to bridge that gap. My goal is to cut through the academic noise and equip you with the hands-on, practical skills you need to build production-ready AI and bring real value to the world. 7+ Years of AI Experience 10k+ Engineers Trained 90k+ AI Community Members 200k+ Followers Across-Platforms WHAT OTHERS ARE SAYING ★★★★★ \"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks. Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming. Highly recommend.\" Pete Huang Co-Founder, The Neuron ★★★★★ \"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance. Very highly recommended!\" Luis Serrano Ph.D., Founder of Serrano Academy ★★★★★ \"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\" Tina Huang Founder of Lonely Octopus, Ex-Meta ★★★★★ \"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\" Sander Schulhoff Founder & CEO of Learn Prompting ★★★★★ \"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\" Alex Volkov Host of ThursdAI.news ★★★★★ \"An incredible survey of all the real-world problems one encounters when trying to productionize an LLM, as well as multiple solutions to each roadblock. Highly recommend this!\" Nick Singh Founder of DataLemur ★★★★★ \"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\" Shashank Kalanithi Data Engineer at Meta ★★★★★ \"This will be valuable to anyone looking to dive into the field quickly and efficiently.\" Jeremy Pinto Senior Applied Research Scientist at Mila ★★★★★ \"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks. Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming. Highly recommend.\" Pete Huang Co-Founder, The Neuron ★★★★★ \"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance. Very highly recommended!\" Luis Serrano Ph.D., Founder of Serrano Academy ★★★★★ \"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\" Tina Huang Founder of Lonely Octopus, Ex-Meta ★★★★★ \"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\" Sander Schulhoff Founder & CEO of Learn Prompting ★★★★★ \"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\" Alex Volkov Host of ThursdAI.news ★★★★★ \"An incredible survey of all the real-world problems one encounters when trying to productionize an LLM, as well as multiple solutions to each roadblock. Highly recommend this!\" Nick Singh Founder of DataLemur ★★★★★ \"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\" Shashank Kalanithi Data Engineer at Meta ★★★★★ \"This will be valuable to anyone looking to dive into the field quickly and efficiently.\" Jeremy Pinto Senior Applied Research Scientist at Mila\n\nStop Making AI Projects That Never Make it Past the Demo\nLearn the exact architecture patterns that separate production-grade AI systems from prototypes. Join 100,000+ engineers upskilling in AI engineering! Get Instant Access", + "links": [], + "chunks": [ + { + "chunk_id": "tai-d324473c51a513fff7b6d760", + "text": "Learn how to make architectural decisions\n\nThe Architecture Mistakes AI Engineers Make\n...and the decision frameworks to prevent them, trusted by engineers at Intel, Europol, Activeloop and more. Get Instant Access arrow_forward Trusted by 100,000+ engineers: ⭐⭐⭐⭐⭐ Louis-François Bouchard Lead AI Engineer & Co-Founder at Towards AI Industry Endorsement \"Helps learners understand everything from fundamentals to advanced building blocks of LLM applications. Most comprehensive resource I've seen.\" Jerry Liu CEO of LlamaIndex Engineers from global leaders are learning with us\n\nAre you making these AI system design mistakes?\nMost AI projects fail in production. Not from lack of skill, but from predictable architectural choices. 01\n\nOver-Engineered AI Systems\nDefaulting to \"agent + tools\" leads to unnecessary complexity, compounding failure points and debugging nightmares. 02\n\nDiscovering Requirements Late\nUnclear requirements cause cost blowups, tool sprawl, and untestable autonomy. 03\n\nReliability Risk\nAI systems fail in production through hallucinated outputs, edge case failures, and poor observability.\n\nWhat You'll Learn\n01\n\nA Process To Make Clean Design Decisions\nInstead of defaulting to agents + tools, you'll have a clear framework to make architectural decisions using constraints. 02\n\nStart With Clear Requirements\nLearn the kickoff questions that force clarity upfront, so cost blowups, tool sprawl, and untestable autonomy don't show up after you've already built half the system. 03\n\nDesign For Reliability From Day 1\nPractical rules for making agents production-safe dialing in a thin agent layer with validation loops, observability, and human-in-the-loop design. 04", + "heading": "The Architecture Mistakes AI Engineers Make", + "index": 0, + "evidence_spans": [ + { + "span_id": "span-bd325a41c1debac96ac63670", + "text": "Learn how to make architectural decisions" + }, + { + "span_id": "span-c6de8da5b5d962fcad2bb19d", + "text": "...and the decision frameworks to prevent them, trusted by engineers at Intel, Europol, Activeloop and more." + }, + { + "span_id": "span-812594a041e59357d58296b0", + "text": "Get Instant Access arrow_forward" + }, + { + "span_id": "span-3b35552d74c466f67263d624", + "text": "Trusted by 100,000+ engineers: ⭐⭐⭐⭐⭐" + }, + { + "span_id": "span-d0d2cc7322f511e87b611fe1", + "text": "Louis-François Bouchard" + }, + { + "span_id": "span-a2fd7306d6b1cd7bfe80b326", + "text": "Lead AI Engineer & Co-Founder at Towards AI" + }, + { + "span_id": "span-f21184f020a62a35fa2a2a96", + "text": "Industry Endorsement" + }, + { + "span_id": "span-38fd00375b6dec18abb53722", + "text": "\"Helps learners understand everything from fundamentals to advanced building blocks of LLM applications." + }, + { + "span_id": "span-467a681d09bc5b302eee5e3f", + "text": "Most comprehensive resource I've seen.\"" + }, + { + "span_id": "span-705962cabe74884af8e106b8", + "text": "Jerry Liu" + }, + { + "span_id": "span-c624eb2a0ea7b5035aa12f84", + "text": "CEO of LlamaIndex" + }, + { + "span_id": "span-2c585515f3929758c8ccb07d", + "text": "Engineers from global leaders are learning with us" + }, + { + "span_id": "span-f7aa40b5e3ba337a0ce897bd", + "text": "Most AI projects fail in production." + }, + { + "span_id": "span-40f3a53872e17ff3f3fca3cd", + "text": "Not from lack of skill, but from predictable architectural choices." + }, + { + "span_id": "span-bcd6aa2f33985e8893ace9b0", + "text": "Defaulting to \"agent + tools\" leads to unnecessary complexity, compounding failure points and debugging nightmares." + }, + { + "span_id": "span-3d5a7b41a7499fa67bfa4cd2", + "text": "Unclear requirements cause cost blowups, tool sprawl, and untestable autonomy." + }, + { + "span_id": "span-2c956bf82cc361768bf1ab6a", + "text": "AI systems fail in production through hallucinated outputs, edge case failures, and poor observability." + }, + { + "span_id": "span-a6379958679fbdf8a3b7cc2e", + "text": "What You'll Learn" + }, + { + "span_id": "span-321cc3eea5a895dd73bc9306", + "text": "Instead of defaulting to agents + tools, you'll have a clear framework to make architectural decisions using constraints." + }, + { + "span_id": "span-5e0950c5bdb718f47d7d9e97", + "text": "Learn the kickoff questions that force clarity upfront, so cost blowups, tool sprawl, and untestable autonomy don't show up after you've already built half the system." + }, + { + "span_id": "span-a2cb013f41078b765f4e430b", + "text": "Design For Reliability From Day 1" + }, + { + "span_id": "span-f83a4324957e0a1ff3d28db1", + "text": "Practical rules for making agents production-safe dialing in a thin agent layer with validation loops, observability, and human-in-the-loop design." + } + ] + }, + { + "chunk_id": "tai-3bf3b3ac37def1f76dd76757", + "text": "Architect For Millions, NOT Your First Demo\nLearn to build the architecture for real user traffic, keeping your agents thin so the system is debuggable and observable in production. Get Instant Access\n\nHi, I'm Louis-François\nLead AI Engineer and Educator", + "heading": "Architect For Millions, NOT Your First Demo", + "index": 1, + "evidence_spans": [ + { + "span_id": "span-83499dab96516a934a9c1c31", + "text": "Learn to build the architecture for real user traffic, keeping your agents thin so the system is debuggable and observable in production." + }, + { + "span_id": "span-3a2135652d2c345579523954", + "text": "Get Instant Access" + }, + { + "span_id": "span-80f5047697a74ea7fde7739d", + "text": "Lead AI Engineer and Educator" + } + ] + }, + { + "chunk_id": "tai-dd70074e81fbbd9e6658bce0", + "text": "Hi, I'm Louis-François\nLead AI Engineer and Educator I’m the co-founder of Towards AI and author of the bestselling book, Building LLMs for Production . After years of research, I dropped out of my Ph.D. because building AI teams in the startup world showed me a harsh reality: traditional universities are failing to prepare engineers for the industry. We created Towards AI to bridge that gap. My goal is to cut through the academic noise and equip you with the hands-on, practical skills you need to build production-ready AI and bring real value to the world. 7+ Years of AI Experience 10k+ Engineers Trained 90k+ AI Community Members 200k+ Followers Across-Platforms WHAT OTHERS ARE SAYING ★★★★★ \"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks. Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming. Highly recommend.\" Pete Huang Co-Founder, The Neuron ★★★★★ \"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance. Very highly recommended!\" Luis Serrano Ph.D., Founder of Serrano Academy ★★★★★ \"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\" Tina Huang Founder of Lonely Octopus, Ex-Meta ★★★★★ \"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\" Sander Schulhoff Founder & CEO of Learn Prompting ★★★★★ \"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\" Alex Volkov Host of ThursdAI.news ★★★★★ \"An incredible survey of all the", + "heading": "Hi, I'm Louis-François", + "index": 2, + "evidence_spans": [ + { + "span_id": "span-ac936a07d118cf64f6749f64", + "text": "Lead AI Engineer and Educator" + }, + { + "span_id": "span-a19ec66bc87e7af0f77ddacf", + "text": "I’m the co-founder of Towards AI and author of the bestselling book, Building LLMs for Production ." + }, + { + "span_id": "span-8ac75c9e4579d5bfcd81c18f", + "text": "After years of research, I dropped out of my Ph.D. because building AI teams in the startup world showed me a harsh reality: traditional universities are failing to prepare engineers for the industry." + }, + { + "span_id": "span-3822188802405fffebde5900", + "text": "We created Towards AI to bridge that gap." + }, + { + "span_id": "span-7d414d8281fcd07cbdffea58", + "text": "My goal is to cut through the academic noise and equip you with the hands-on, practical skills you need to build production-ready AI and bring real value to the world." + }, + { + "span_id": "span-430d540a60f4a69bff30c8a9", + "text": "Years of AI" + }, + { + "span_id": "span-11b8baf4d4b0f220702c737a", + "text": "AI Community" + }, + { + "span_id": "span-2235693d0b0f7ba34b677846", + "text": "Across-Platforms" + }, + { + "span_id": "span-3a7c8db171bf2ad17b9fc759", + "text": "WHAT OTHERS ARE SAYING" + }, + { + "span_id": "span-56ca7c73d5f5e28b1d60c28e", + "text": "\"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks." + }, + { + "span_id": "span-108865a8b797c00c819ca9c9", + "text": "Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming." + }, + { + "span_id": "span-97b8cd16840454590aac6eb2", + "text": "Highly recommend.\"" + }, + { + "span_id": "span-a5a35279015e527cec244091", + "text": "Pete Huang" + }, + { + "span_id": "span-3188d3a8b60cec2409ef114e", + "text": "Co-Founder, The Neuron" + }, + { + "span_id": "span-b52e3191c1f99d43132e278c", + "text": "\"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance." + }, + { + "span_id": "span-c58f9d9797f5fc26879f3e11", + "text": "Very highly recommended!\"" + }, + { + "span_id": "span-dc4d61bf107dd788917287d0", + "text": "Luis Serrano" + }, + { + "span_id": "span-bd3ba81dddb82b713896714c", + "text": "Ph.D., Founder of Serrano Academy" + }, + { + "span_id": "span-c5c691112b319f853286e458", + "text": "\"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\"" + }, + { + "span_id": "span-8475b491923908793a525ef3", + "text": "Tina Huang" + }, + { + "span_id": "span-88564c4cc582b03bcd4b5eb1", + "text": "Founder of Lonely Octopus, Ex-Meta" + }, + { + "span_id": "span-cdf0109beb26af8bd05be177", + "text": "\"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\"" + }, + { + "span_id": "span-c58a8b166a7a59ae0b41f8d4", + "text": "Sander Schulhoff" + }, + { + "span_id": "span-a3d79f7eae3d83ba03718d73", + "text": "Founder & CEO of Learn Prompting" + }, + { + "span_id": "span-fc590fe456589bd970796d02", + "text": "\"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\"" + }, + { + "span_id": "span-a48298f71c910a1a74eef470", + "text": "Alex Volkov" + }, + { + "span_id": "span-086b795b8614c060f9c16914", + "text": "Host of ThursdAI.news" + } + ] + }, + { + "chunk_id": "tai-e00078586dd147b44e7f4fca", + "text": "Hi, I'm Louis-François\nreal-world problems one encounters when trying to productionize an LLM, as well as multiple solutions to each roadblock. Highly recommend this!\" Nick Singh Founder of DataLemur ★★★★★ \"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\" Shashank Kalanithi Data Engineer at Meta ★★★★★ \"This will be valuable to anyone looking to dive into the field quickly and efficiently.\" Jeremy Pinto Senior Applied Research Scientist at Mila ★★★★★ \"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks. Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming. Highly recommend.\" Pete Huang Co-Founder, The Neuron ★★★★★ \"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance. Very highly recommended!\" Luis Serrano Ph.D., Founder of Serrano Academy ★★★★★ \"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\" Tina Huang Founder of Lonely Octopus, Ex-Meta ★★★★★ \"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\" Sander Schulhoff Founder & CEO of Learn Prompting ★★★★★ \"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\" Alex Volkov Host of ThursdAI.news ★★★★★ \"An incredible survey of all the real-world problems one encounters when trying to productionize an LLM, as well as multiple solutions to each roadblock. Highly recommend this!\" Nick", + "heading": "Hi, I'm Louis-François", + "index": 3, + "evidence_spans": [ + { + "span_id": "span-686a65004b391c33737589a6", + "text": "\"A truly wonderful resource that develops understanding of LLMs from the ground up, from theory to code and modern frameworks." + }, + { + "span_id": "span-b8facf916f2f43e0eb810aa0", + "text": "Grounds your knowledge in research trends and frameworks that develop your intuition around what's coming." + }, + { + "span_id": "span-e14552af345ec196a8ca3d4c", + "text": "Highly recommend.\"" + }, + { + "span_id": "span-681650cca240e69c2196dfc1", + "text": "Pete Huang" + }, + { + "span_id": "span-57e6224e32568638a04a459d", + "text": "Co-Founder, The Neuron" + }, + { + "span_id": "span-5522498bd99f1c49131fc5b8", + "text": "\"Contains thorough explanations and code for you to start using and deploying LLMs, as well as optimizing their performance." + }, + { + "span_id": "span-371897fd93e74acab281b71c", + "text": "Very highly recommended!\"" + }, + { + "span_id": "span-e164fa69351530ad076531ac", + "text": "Luis Serrano" + }, + { + "span_id": "span-390798f8a323a4bee3f863c2", + "text": "Ph.D., Founder of Serrano Academy" + }, + { + "span_id": "span-6e9bb6d8f8ceb608a23adcc2", + "text": "\"A comprehensive and well-rounded resource that covers all the fundamentals of LLMs with a well-struck balance between theory and code.\"" + }, + { + "span_id": "span-3f952376fb2d9e24af1b637f", + "text": "Tina Huang" + }, + { + "span_id": "span-dd37ec5f29ecdd7c43f87c28", + "text": "Founder of Lonely Octopus, Ex-Meta" + }, + { + "span_id": "span-b5e1794532df39a9f614f535", + "text": "\"Towards AI has done a great job assembling all of the technical resources needed by a modern GenAI applied practitioner.\"" + }, + { + "span_id": "span-ff7d86d62d60defc95386bcc", + "text": "Sander Schulhoff" + }, + { + "span_id": "span-e28a6415e579fbe32dce8fd4", + "text": "Founder & CEO of Learn Prompting" + }, + { + "span_id": "span-09e2493746db539d1c7df0ee", + "text": "\"Louis and the Towards AI team have created an essential resource for developers who want to expand their AI expertise and apply it to real-world challenges.\"" + }, + { + "span_id": "span-073c595867f676be3c9613d6", + "text": "Alex Volkov" + }, + { + "span_id": "span-c2c2ff31de0e2fd747f028e8", + "text": "Host of ThursdAI.news" + }, + { + "span_id": "span-b0d92f3b7f6bd97f71d39be3", + "text": "\"An incredible survey of all the real-world problems one encounters when trying to productionize an LLM, as well as multiple solutions to each roadblock." + }, + { + "span_id": "span-efe40ddccdfd192a2203f105", + "text": "Highly recommend this!\"" + }, + { + "span_id": "span-fd661db5acd3f2f0256cea74", + "text": "Nick Singh" + }, + { + "span_id": "span-44beb70091136cc118c19506", + "text": "Founder of DataLemur" + }, + { + "span_id": "span-cd24e997d7fcca5049c008f9", + "text": "\"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\"" + }, + { + "span_id": "span-64d7f4e80e0aaf5a433a6a8b", + "text": "Shashank Kalanithi" + }, + { + "span_id": "span-8b61e78939406eba7951e48b", + "text": "Data Engineer at Meta" + }, + { + "span_id": "span-38c0d7217f9b2121f125422b", + "text": "\"This will be valuable to anyone looking to dive into the field quickly and efficiently.\"" + }, + { + "span_id": "span-7c3ac6b832d6a5e373f00a76", + "text": "Jeremy Pinto" + }, + { + "span_id": "span-de4dde6418e762a21efd875f", + "text": "Senior Applied Research Scientist at Mila" + } + ] + }, + { + "chunk_id": "tai-a1b0124fe768390f1dc54bd6", + "text": "Hi, I'm Louis-François\nSingh Founder of DataLemur ★★★★★ \"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\" Shashank Kalanithi Data Engineer at Meta ★★★★★ \"This will be valuable to anyone looking to dive into the field quickly and efficiently.\" Jeremy Pinto Senior Applied Research Scientist at Mila\n\nStop Making AI Projects That Never Make it Past the Demo\nLearn the exact architecture patterns that separate production-grade AI systems from prototypes. Join 100,000+ engineers upskilling in AI engineering! Get Instant Access", + "heading": "Hi, I'm Louis-François", + "index": 4, + "evidence_spans": [ + { + "span_id": "span-f3ff3fafabb25ca81a084c82", + "text": "Founder of DataLemur" + }, + { + "span_id": "span-c0e89eef167702cfa697f02b", + "text": "\"An indispensable guide for anyone venturing into the world of large language models...Covering everything from theory to practical deployment.\"" + }, + { + "span_id": "span-44eab35d3a2da0445870a2b8", + "text": "Shashank Kalanithi" + }, + { + "span_id": "span-f512d91f1a7afb45a3a16edd", + "text": "Data Engineer at Meta" + }, + { + "span_id": "span-19a827051bf3365a3fdc7149", + "text": "\"This will be valuable to anyone looking to dive into the field quickly and efficiently.\"" + }, + { + "span_id": "span-387c43deeeb4788da7bc34f6", + "text": "Jeremy Pinto" + }, + { + "span_id": "span-149d4aba501c904afed2e695", + "text": "Senior Applied Research Scientist at Mila" + }, + { + "span_id": "span-27eb0a0f62139132b6082b1b", + "text": "Learn the exact architecture patterns that separate" + }, + { + "span_id": "span-dcdefb2a98e0cc2e99a8bf31", + "text": "production-grade AI systems from prototypes." + }, + { + "span_id": "span-b7055caa6702705ea8973224", + "text": "Join 100,000+ engineers upskilling in AI engineering!" + }, + { + "span_id": "span-e3daf966d3257e5c4d9cb6e0", + "text": "Get Instant Access" + } + ] + } + ], + "fetched_at": "2026-08-09T11:24:43.395909+00:00", + "content_sha256": "463aac2b3baa5a120c1a22ff40d009db48b9a8a0f3cba7ed5b1f11df35697a18", + "content_hash": "463aac2b3baa5a120c1a22ff40d009db48b9a8a0f3cba7ed5b1f11df35697a18", + "authority": "official_site", + "status": "included", + "retrieval_eligible": true, + "http_status": 200, + "lastmod": "", + "sitemap_url": "https://towardsai.com/pages-sitemap.xml", + "redirect_chain": [ + "https://towardsai.com/webinars/agentengineering/" + ], + "evidence_hash": "fb58cc78dd1792c0c5b35fc467170cc20e71ea441e9a4edf101752a5e5953412" + } + ] +}