Rentelo FAQ AI (End-to-End Project)

An offline, self-contained FAQ chatbot for Rentelo that gives correct answers to your policy/FAQ questions using semantic search (TF‑IDF cosine similarity)—no external APIs or keys required.

What you get

• backend/: FastAPI app with a local FAQ knowledge base and semantic search.
• frontend/: Vite + React + Tailwind chat UI (clean, responsive).
• docker-compose.yml: One command to run both services.

Run locally (no Docker)

1) Backend

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

pip install -r requirements.txt
uvicorn app:app --host 0.0.0.0 --port 8000

Backend will be on http://localhost:8000

2) Frontend

Open a second terminal:

cd frontend
npm install
npm run dev

Frontend will be on http://localhost:5173 (Vite default). It calls the backend at http://localhost:8000.

Run with Docker

docker compose up --build

• Frontend: http://localhost:3000
• Backend: http://localhost:8000

API

• POST /ask — Body: { "question": "Your text" } → returns best matching answer and metadata.
• GET /faqs — List all FAQ entries (title, tags).

How it works

• The backend loads FAQ data from data/faqs.json (Q/A + tags).
• Builds a TF‑IDF vector space using both questions and answers to improve recall.
• Computes cosine similarity between your query and the knowledge base.
• Returns the top answer. (Optional: supports top_k.)

Edit/extend FAQs

• Update backend/data/faqs.json with new entries (fields: id, title, question, answer, tags list).
• Restart the backend to reindex.

---

Made for the requirements you shared (refund policy, documents, KYC, timings, etc.).