# 🧠 ResumeMatch AI

A smart, dual-purpose machine learning project that empowers **job seekers** and **recruiters** with intelligent, automated resume-job matching.

---

## 🚀 Overview

**ResumeMatch AI** is an ML-powered system that does two things:

- 🎯 Helps **candidates** find the most relevant job listings based on their resume.
- 🧑‍💼 Helps **recruiters** rank and filter a pool of resumes based on how well they match a specific job description.

Whether you're applying or hiring, ResumeMatch AI works both ways, all powered by a unified vector similarity engine under the hood.

---

## 🧩 Problem Statement

Traditional job applications are inefficient — resumes are often overlooked, and job seekers apply blindly to roles. Recruiters, on the other hand, must manually sift through large stacks of resumes to find a match.

**This project solves both problems:**
- For job seekers: It suggests the most suitable job listings.
- For employers: It filters and ranks resumes based on relevance to a job posting.

---

### 🔄 Dual Functionality

- **Job Seeker Mode**  
  - **Input:** Resume (PDF or text)  
  - **Output:** Ranked list of matching job descriptions based on skills, experience, and keywords

- **Recruiter Mode**  
  - **Input:** Job description  
  - **Output:** Ranked list of candidate resumes that best fit the role

## 📁 Dataset Not Included

This repo does **not** include the `Resume.csv` dataset due to file size limits.

To use this project:
1. Download the datasets manually:
   ↳ Resume dataset used for training [Kaggle](https://www.kaggle.com/datasets/snehaanbhawal/resume-dataset)
   ↳ Job dataset used for training [Kaggle](https://www.kaggle.com/datasets/ravindrasinghrana/job-description-dataset?select=job_descriptions.csv)
2. Place resume dataset in the `data/raw/resumes` directory and the jobs dataset in `data/raw/job_descriptions`.
3. You're good to go!