AI Chatbot

An open-source, local-first Retrieval-Augmented Generation (RAG) chatbot built with LangChain, Groq, and Streamlit.

It enables you to chat with your custom documentation (e.g., PostHog Handbook) using state-of-the-art LLMs like Llama 3 for free, with a focus on privacy and speed.

✨ Features

• ⚡ Blazing Fast Generation: Powered by Groq API (Llama 3 70B) for near-instant responses.
• 🔍 Hybrid Search: Combines Semantic Search (ChromaDB + Sentence Transformers) and Keyword Search (BM25) for high-precision retrieval.
• 📄 Document Ingestion: Automatically parses and chunks Markdown/text documentation.
• 💬 Conversational UI: Clean, responsive interface built with Streamlit.
• 🛡️ Privacy-First: Runs locally (except for the LLM API call), keeping your vector index on your machine.
• 📚 Verifiable Answers: Citations and source references for every response.

🚀 Quick Start

Prerequisites

• Python 3.10+
• Groq API Key (Free)

Installation

1. Clone the repository

   git clone https://github.com/yourusername/rag-chatbot.git
   cd rag-chatbot

2. Create a virtual environment

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

3. Install dependencies

   pip install -r requirements.txt

4. Configure Environment Create a .env file in the root directory:

   # .env
   GROQ_API_KEY=your_groq_api_key_here
   
   # Configuration
   CHUNK_SIZE=500
   CHUNK_OVERLAP=50
   TOP_K=5
   EMBEDDING_MODEL=sentence-transformers/all-MiniLM-L6-v2

Usage

1. Ingest Data (First run only) Place your markdown files in data/raw or use the default loader.

   python src/ingestion/loader.py

2. Run the App

   python -m streamlit run src/ui/app.py

   Your browser should open at http://localhost:8501.

📂 Documentation

• Architecture Overview: Deep dive into the RAG pipeline and system design.
• API Reference: Details on the internal modules.

🛠️ Tech Stack

• LLM Provider: Groq (Llama-3.3-70b-versatile)
• Orchestration: LangChain
• Vector Store: ChromaDB
• Embeddings: sentence-transformers/all-MiniLM-L6-v2
• Retrieval: rank_bm25 (Sparse) + Chroma (Dense) -> Ensemble
• UI: Streamlit

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.