Job Finder

A portfolio-grade job recommendation system that uses NLP/ML to match your resume to job postings with explainable recommendations.

Features

• Resume Parsing: Upload your resume (PDF/TXT) and extract skills automatically
• Semantic Matching: Uses sentence-transformers to find jobs that match your experience
• Hybrid Scoring: Combines embedding similarity, skill overlap, location, salary, and recency
• Explainable Recommendations: See why each job was recommended
• Modern React UI: Fast, responsive interface with real-time updates
• Local-first: Runs entirely on your machine with persistent data

Tech Stack

Component	Technology
Backend	FastAPI (Python 3.11+)
Frontend	React 19 + TypeScript + Vite
NLP Model	all-MiniLM-L6-v2 (sentence-transformers)
Vector DB	ChromaDB
Metadata DB	SQLite
Data Source	Adzuna API
Resume Parsing	pdfplumber

Quick Start

Option 1: Automated Setup (Recommended)

git clone https://github.com/BillSteinUNB/JobFinder.git
cd JobFinder
python scripts/setup_local.py

Option 2: Manual Setup

1. Create Python environment:

python -m venv .venv
.venv\Scripts\activate  # Windows
pip install -r requirements.txt

2. Install frontend:

cd jobhunt && npm install && cd ..

3. Configure environment - create .env:

ADZUNA_APP_ID=your_app_id
ADZUNA_APP_KEY=your_app_key

4. Collect jobs and build index:

python scripts/collect_all_jobs.py --target 1000
python scripts/build_vector_index.py

Running the App

Terminal 1 - Backend:

uvicorn api.main:app --reload --port 8000

Terminal 2 - Frontend:

cd jobhunt && npm run dev

Open: http://localhost:5173

Public Access with ngrok

For demos/interviews:

1. Install ngrok from https://ngrok.com/download
2. Start backend: uvicorn api.main:app --reload --port 8000
3. Expose backend: ngrok http 8000
4. Update jobhunt/.env.local with ngrok URL
5. Restart frontend

License

MIT