Update hharsha/agentic-systems-showcase

README.md

@@ -6,6 +6,9 @@ tags:
   - agents
   - rag
   - llmops
+  - multi-agent
+  - mcp
+  - portfolio
 pretty_name: Agentic Systems Showcase
 size_categories:
   - n<1K
@@ -18,9 +21,11 @@ configs:
 
 # Agentic Systems Showcase
 
-Short records for public projects by **Hanumanthu Harsha Vardhan** ([GitHub hharsha98](https://github.com/hharsha98), Hub [hharsha](https://huggingface.co/hharsha)).
+Catalog of **public** projects by **Hanumanthu Harsha Vardhan**
+([GitHub `hharsha98`](https://github.com/hharsha98), Hub [`hharsha`](https://huggingface.co/hharsha)).
 
-Each row is a name, a URL, and a one-paragraph summary taken from the project's own README or from the Space that ships with this profile. There are **no papers, no benchmarks invented for this card, and no third-party repos**.
+Each row is an honest summary taken from the project's own README, Space card, or studio site.
+**No invented papers or benchmarks.**
 
 | Field | Meaning |
 | --- | --- |
@@ -28,14 +33,21 @@ Each row is a name, a URL, and a one-paragraph summary taken from the project's
 | `name` | Display name |
 | `kind` | `github`, `space`, or `website` |
 | `url` | Canonical link |
-| `summary` | What it is, in one paragraph |
-| `stack` | Coarse tags |
+| `summary` | One-paragraph description |
+| `stack` | Coarse tech/topic tags |
 | `owner` | GitHub or Hub handle |
 
-Studio site: https://agentic-systems-studio.com
-Location called out on the profile: Nuremberg, Germany.
-Cloud scope called out on the profile: AWS and GCP.
+**Rows:** 16 (expanded from an earlier 9-row seed).
 
-## How this was written
+## Highlights
 
-Summaries were paraphrased from the public GitHub READMEs of `agentfleet`, `Vibespace`, `agent-os`, `retrievallab`, `agentops-studio`, and `careeragent` in September 2026, plus the two Spaces published with this dataset. If a README later changes, this file can drift; the URL is the source of truth.
+- **Control Tower** — RevenueOps supervisor/worker multi-agent graph with RAG + evals
+- **agentgrid** / **Vibespace** — browser/desktop agentic IDEs (terminal grids, kanban)
+- **AgentFleet** / **AgentOps Studio** — self-hostable multi-agent ops + IaC scaffolding
+- **RetrievalLab** (+ Spaces) — hybrid search, rerank, recall@k / MRR
+- **CareerAgent** — job-hunt agents with cited CV chat
+- Studio site: [agentic-systems-studio.com](https://agentic-systems-studio.com)
+
+Related training data for tag generation: [`hharsha/agentic-github-meta`](https://huggingface.co/datasets/hharsha/agentic-github-meta).
+
+Location: Nuremberg, Germany. Cloud scope: AWS and GCP.

data/projects.jsonl

@@ -1,9 +1,16 @@
-{"id": "agentfleet", "name": "AgentFleet", "kind": "github", "url": "https://github.com/hharsha98/agentfleet", "summary": "Self-hostable multi-agent operations platform: tool-using agents, a goal run as a task DAG, evals, cost governance, and guardrails.", "stack": ["python", "agents", "llmops"], "owner": "hharsha98"}
-{"id": "vibespace", "name": "Vibespace", "kind": "github", "url": "https://github.com/hharsha98/Vibespace", "summary": "Open-source agentic development environment with a terminal grid, a kanban that dispatches coding agents, and shared agent memory.", "stack": ["typescript", "agents", "mcp"], "owner": "hharsha98"}
-{"id": "agent-os", "name": "Agent OS", "kind": "github", "url": "https://github.com/hharsha98/agent-os", "summary": "Local-first command center for coding agents already on the machine. Dry-run by default, sandboxed workspace, gated machine control.", "stack": ["javascript", "agents", "local-first"], "owner": "hharsha98"}
-{"id": "retrievallab", "name": "RetrievalLab", "kind": "github", "url": "https://github.com/hharsha98/retrievallab", "summary": "Advanced RAG made visible: contextual chunking, hybrid search, cross-encoder reranking, query transformation, and retrieval eval (recall@k, MRR).", "stack": ["typescript", "rag", "evaluation"], "owner": "hharsha98"}
-{"id": "agentops-studio", "name": "AgentOps Studio", "kind": "github", "url": "https://github.com/hharsha98/agentops-studio", "summary": "Portable multi-agent operations lab: orchestration, RAG with citations, MCP-style tool registry, traces, and optional container scaffolding. Separate from AgentFleet.", "stack": ["python", "agents", "llmops", "mcp"], "owner": "hharsha98"}
-{"id": "careeragent", "name": "CareerAgent", "kind": "github", "url": "https://github.com/hharsha98/careeragent", "summary": "Job-hunt agents: cited chat over a CV, company research, CV tailoring, and a Kanban tracker.", "stack": ["python", "rag", "agents"], "owner": "hharsha98"}
-{"id": "retrieval-lab-demo", "name": "Retrieval Lab Demo", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/retrieval-lab-demo", "summary": "Free CPU Gradio demo: upload a PDF, chunk, embed with MiniLM (ONNX), retrieve top-k, answer with citations.", "stack": ["gradio", "rag", "cpu"], "owner": "hharsha"}
-{"id": "agentic-systems-studio-space", "name": "Agentic Systems Studio Space", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/agentic-systems-studio", "summary": "Landing cards for the studio plus a plan-tools-answer agent trace that runs on CPU.", "stack": ["gradio", "agents", "cpu"], "owner": "hharsha"}
-{"id": "studio-site", "name": "Agentic Systems Studio", "kind": "website", "url": "https://agentic-systems-studio.com", "summary": "Public studio site for Hanumanthu Harsha Vardhan.", "stack": ["web"], "owner": "hharsha98"}
+{"id": "agent-os", "name": "Agent OS", "kind": "github", "url": "https://github.com/hharsha98/agent-os", "summary": "Local-first command center for coding agents already on the machine (Cursor, Claude, Codex, Hermes). Dry-run by default, sandboxed workspace, gated machine control. React/TypeScript dashboard.", "stack": ["javascript", "typescript", "react", "agents", "local-first"], "owner": "hharsha98"}
+{"id": "agentfleet", "name": "AgentFleet", "kind": "github", "url": "https://github.com/hharsha98/agentfleet", "summary": "Self-hostable multi-agent operations platform: chat with tool-using agents, hand the orchestrator a goal and watch it execute as a live task DAG, with production-grade evals, cost governance, and guardrails.", "stack": ["python", "fastapi", "agents", "multi-agent", "llmops", "evals", "docker"], "owner": "hharsha98"}
+{"id": "agentgrid", "name": "agentgrid", "kind": "github", "url": "https://github.com/hharsha98/agentgrid", "summary": "BridgeSpace-inspired agentic development environment — mission control for running multiple AI coding agents side by side in one browser window. Cursor lane, intentionally separate from Vibespace/vibedeck.", "stack": ["typescript", "nextjs", "agents", "multi-agent", "developer-tools"], "owner": "hharsha98"}
+{"id": "agentops-studio", "name": "AgentOps Studio", "kind": "github", "url": "https://github.com/hharsha98/agentops-studio", "summary": "Portable multi-agent operations lab: orchestration, RAG with citations, MCP-style tool registry, traces/observability, Docker/Kubernetes and Terraform scaffolding. Separate from AgentFleet; focused on business ops portability.", "stack": ["python", "agents", "llmops", "mcp", "rag", "docker", "kubernetes", "terraform"], "owner": "hharsha98"}
+{"id": "ai-rag-project", "name": "RAG Chat with Documents", "kind": "github", "url": "https://github.com/hharsha98/ai-rag-project", "summary": "Chat with your PDFs: ingestion, embeddings, ChromaDB retrieval with citations, Streamlit UI, and an LLM-as-judge evaluation suite for answer quality.", "stack": ["python", "streamlit", "rag", "chromadb", "evaluation"], "owner": "hharsha98"}
+{"id": "automaton", "name": "Automaton", "kind": "github", "url": "https://github.com/hharsha98/automaton", "summary": "Experimental sovereign-agent concept: agents that can provision compute, register domains, and fund their own runtime — exploring self-improving, self-replicating agent economics (research/portfolio, not a production product).", "stack": ["python", "agents", "experimental"], "owner": "hharsha98"}
+{"id": "careeragent", "name": "CareerAgent", "kind": "github", "url": "https://github.com/hharsha98/careeragent", "summary": "Job-hunt multi-agent platform: cited chat over a CV, live company research, CV tailoring anchored to evidence, Kanban application tracker, with cost metering. FastAPI + React + pgvector; deployed on HF Spaces + Cloudflare Pages.", "stack": ["python", "fastapi", "react", "rag", "agents", "pgvector"], "owner": "hharsha98"}
+{"id": "control-tower", "name": "RevenueOps Agent Control Tower", "kind": "github", "url": "https://github.com/hharsha98/06-revenue-ops-agent-control-tower", "summary": "Enterprise-style RevenueOps multi-agent AI control tower for founders and COOs. A supervisor agent coordinates specialists for sales, support, customer communication, engineering handoff, risk checks, evals, and audit trails — with RAG and live tools.", "stack": ["python", "langgraph", "agents", "multi-agent", "rag", "evals", "observability"], "owner": "hharsha98"}
+{"id": "portfolio", "name": "Agentic Systems Studio Portfolio", "kind": "github", "url": "https://github.com/hharsha98/portfolio", "summary": "Hand-built portfolio site for shipped RAG and multi-agent AI systems — live labs, local-first tools, and products. React/TypeScript front end for Agentic Systems Studio.", "stack": ["typescript", "react", "nextjs", "web"], "owner": "hharsha98"}
+{"id": "retrievallab", "name": "RetrievalLab", "kind": "github", "url": "https://github.com/hharsha98/retrievallab", "summary": "Advanced RAG made visible: contextual chunking, hybrid search, cross-encoder reranking, query transformation, and retrieval eval (recall@k, MRR). FastAPI + React + pgvector, designed for €0/mo hosting.", "stack": ["typescript", "fastapi", "react", "rag", "pgvector", "evaluation"], "owner": "hharsha98"}
+{"id": "vibespace", "name": "Vibespace", "kind": "github", "url": "https://github.com/hharsha98/Vibespace", "summary": "Open-source agentic development environment: run Claude Code, cursor-agent, Codex and shells side by side in a terminal grid, with a kanban board that dispatches them, shared agent memory, and multi-agent orchestration (TypeScript/Tauri).", "stack": ["typescript", "tauri", "agents", "multi-agent", "mcp", "developer-tools"], "owner": "hharsha98"}
+{"id": "agent-portfolio-space", "name": "Agent Portfolio Space", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/agent-portfolio", "summary": "Gradio Space portfolio overview of agentic and RAG projects by hharsha — cards, links, and short demos for industrial AI / LLMOps themes.", "stack": ["gradio", "agents", "rag", "portfolio"], "owner": "hharsha"}
+{"id": "agentic-systems-studio-space", "name": "Agentic Systems Studio Space", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/agentic-systems-studio", "summary": "Static/Gradio landing for Agentic Systems Studio: project cards plus a small plan-tools-answer agent trace that runs on free CPU.", "stack": ["gradio", "agents", "cpu", "web"], "owner": "hharsha"}
+{"id": "retrieval-lab-demo", "name": "Retrieval Lab Demo", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/retrieval-lab-demo", "summary": "Free CPU Gradio/static demo: upload a PDF, chunk, embed with MiniLM (ONNX), retrieve top-k, answer with citations. Companion to RetrievalLab.", "stack": ["gradio", "rag", "cpu", "onnx", "embeddings"], "owner": "hharsha"}
+{"id": "retrievallab-space", "name": "RetrievalLab Space", "kind": "space", "url": "https://huggingface.co/spaces/hharsha/retrievallab", "summary": "Gradio Space companion for RetrievalLab: experiment with hybrid search, reranking, and retrieval metrics in a browser without PRO GPU.", "stack": ["gradio", "rag", "hybrid-search", "reranking"], "owner": "hharsha"}
+{"id": "studio-site", "name": "Agentic Systems Studio", "kind": "website", "url": "https://agentic-systems-studio.com", "summary": "Public studio website for Hanumanthu Harsha Vardhan (Nuremberg): live labs, local-first tools, and shipping products across agents, RAG, and LLMOps on AWS/GCP.", "stack": ["web", "nextjs", "agents", "rag", "llmops"], "owner": "hharsha98"}

data/projects.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:848e4640ab8fad5866ff0c2fbbebb73d6621d64d901427837a3d02e5aa92480d
+size 8510