RAG Based Chat Bot

A chatbot powered by Retrieval-Augmented Generation (RAG) using LangChain and Groq.

---

📌 Abstract

This project demonstrates how Retrieval-Augmented Generation (RAG) can be used to create a chatbot that retrieves relevant information from documents, remembers conversations, and responds naturally. It integrates LangChain, ChromaDB, and Groq for efficient document search and fast inference.

---

⚡ Features

• 🔎 Document Retrieval – Fetches context from text files.
• 🧠 Conversation Memory – Remembers last 5 exchanges.
• 📝 Custom Prompting – Controlled responses via templates.
• 📚 Chroma Vector DB – Embedding storage.
• 🎨 Gradio UI – Simple web-based chatbot interface.

---

🛠️ Methodology / How It Works

1. User enters a query.
2. The chatbot uses embeddings (all-MiniLM-L6-v2) to search the vector database.
3. Relevant documents are retrieved.
4. LangChain constructs a response using RAG and Groq.
5. Answer is displayed in the Gradio UI.

---

📂 Project Structure

├── src/           # main chatbot code
├── data/          # text documents for knowledge base
├── chroma_db/     # vector database
├── requirements.txt
├── .env.example   # environment variables
└── README.md

---

🚀 Setup & Installation

# 1. Clone the repository
git clone https://github.com/hasabirforever/rag-chatbot.git
cd rag-chatbot

# 2. Create virtual environment
python -m venv venv
source venv/bin/activate   # Mac/Linux
venv\Scripts\activate      # Windows

# 3. Install dependencies
pip install -r requirements.txt

# 4. Add API keys
cp .env.example .env
# Edit with your GROQ_API_KEY

---

📊 Results / Demo

• Works with knowledge base text files placed in data/.
• Can be extended to support PDFs, docs, or external APIs.
• Provides relevant, context-aware responses.

---

📎 Links

• GitHub Repository: rag-chatbot