Enterprise RAG Chatbot (Baseline)

This repo contains a baseline FAQ-retrieval chatbot:

• Backend: FastAPI + sentence-transformers embeddings + cosine similarity
• Frontend: simple static index.html that calls the backend

Next step (Group project): upgrade to LLM-powered RAG with private document ingestion + access control + cloud deployment.

---

Run locally (Windows / Conda)

1) Create & activate environment (first time only)

conda create -n chatbot-local python=3.12 -y
conda activate chatbot-local

2) Install dependencies

cd backend
pip install -r requirements.txt

3) Build embeddings (required)

cd backend
python embedding_builder.py

4) Start backend

cd backend
python -m uvicorn main:app --host 127.0.0.1 --port 8000 --reload

Open:

• http://127.0.0.1:8000/docs
• http://127.0.0.1:8000/health

5) Start frontend

cd frontend
python -m http.server 5500

Open:

• http://127.0.0.1:5500/index.html

---

Troubleshooting

• If a new terminal opens in (base), run:

conda activate chatbot-local

• If you see ModuleNotFoundError, reinstall:

pip install -r backend/requirements.txt

• If you changed faqs.json, rebuild embeddings:

python backend/embedding_builder.py

---

Repo notes

Generated files are not meant to be committed:

• backend/index.npy
• backend/meta.json

---

Also do this before pushing

Make sure you have a .gitignore so you don’t upload generated files + cache:

C:\Git\cloud-rag-chatbot\.gitignore

__pycache__/
*.pyc
.env
.vscode/

backend/index.npy
backend/meta.json