LexRetriever — Fullstack (React + FastAPI)

This repo is organized into two main folders:

• frontend/ — React + Vite UI
• backend/ — FastAPI API, Pinecone + HuggingFace integration, and dataset builders
• docs/ — reports, archives, and other supporting artifacts

---

Frontend (React + Vite)

Install dependencies

cd frontend
npm install

Run in development mode

npm run dev

The dev server runs at http://localhost:3000 and proxies /api/* to http://localhost:8000.

Build for production

npm run build

The build output will appear in frontend/dist/. The backend is configured to serve this build automatically when running from backend/.

---

Backend (FastAPI)

Setup

cd backend
python -m venv .venv
# Windows PowerShell
.\.venv\Scripts\Activate.ps1
# macOS / Linux
# source .venv/bin/activate

pip install -r requirements.txt

Run

uvicorn main:app --reload --port 8000

Notes

• The backend serves the frontend build from ../frontend/dist/ when it exists.
• API endpoints are under /api/* (e.g., /api/search, /api/chat).

---

Building/Updating the Vector DBs

The backend includes scripts to build Pinecone indexes from local datasets:

• build_pinecone_judgements.py — build/query vector DB for judgement documents
• build_pinecone_legal.py — build/query vector DB for legal framework sources
• build_legal_rag.py — build a local Chroma DB for constitution+law content

Run these from the backend/ directory.

---

Directory Layout

frontend/      # React app (Vite)
backend/       # FastAPI server + Pinecone/LLM tooling + data
  ├── data/    # datasets, models, and indexes
  ├── main.py
  ├── requirements.txt
  └── ...
docs/          # reports and archive artifacts