🐍 Python RAG Chat (Flask + Gemini + FAISS)

A simple and elegant Retrieval-Augmented Generation (RAG) web app built with Flask, FAISS, and Google Gemini.
It allows you to chat with a local knowledge base of Python documentation, cheatsheets, and curated open-source libraries — all retrieved using semantic search.

---

Features

• Semantic Search using FAISS + sentence-transformers embeddings
• Context-aware Answers powered by Google Gemini (free-tier compatible)
• Python Knowledge Base built from Markdown docs and public datasets
• Lightweight Flask App with clean HTML + CSS frontend
• Local RAG System — no external vector DB needed

---

📚 Data Sources

This RAG system is built using data from:

1. Python Cheatsheet — general Python syntax, functions, modules, and usage examples.
2. Hugging Face: dylanhogg/awesome-python Dataset — curated collection of Python projects and libraries with descriptions and metadata.

Both sources are used locally to create FAISS embeddings for semantic retrieval.

---

⚙️ Setup Instructions

Clone the repository

git clone <your-repo-url>
cd python-rag-chat

Create a virtual environment

Highly recomended to use venv:

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

Install dependencies

pip install -r requirements.txt

Add your Gemini API key

GOOGLE_API_KEY=your_free_tier_gemini_key_here

The free-tier Gemini API works perfectly for this project.

Build your FAISS index

Run this once to generate embeddings from your Markdown and JSON data:

python build_rag_index.py

This will create a FAISS index inside embeddings/faiss_index/. Note : its included in the repository so its optional.

Run the Flask RAG app

python app.py

---