🤖 RAG-Based AI Chatbot 🔥

A Retrieval-Augmented Generation (RAG) AI Chatbot that enhances responses by retrieving relevant documents from an embedded Pinecone vector database before generating intelligent answers using LLMs.

🚀 Features

✅ FastAPI Backend for efficient API-based interactions
✅ Pinecone Vector Database for document retrieval
✅ LangChain Integration for structured prompt execution
✅ Document Embedding & Storage (Fetch, Process, Store)
✅ Question Rewriting for Better Search Queries
✅ Hallucination & Answer Grading System
✅ Dockerized for Easy Deployment

---

📂 Project Structure

rag-chatbot/
│── app/                           # FastAPI application
│   ├── routes/                    # API routes
│   │   ├── rag.py                  # Main RAG API route
│   ├── workflows/                  # Core RAG logic
│   │   ├── nodes.py                 # Document retrieval, grading, generation
│   │   ├── edges.py                 # Workflow decision-making
│   │   ├── graphs.py                # Workflow definition
│   ├── scripts/
│   │   ├── embedding_service.py     # Pinecone embedding & storage
│   ├── dtos/                         # Request & response schemas
│   │   ├── rag.py
│   ├── agent.py                      # LLM & prompt configurations
│── data/                              # Document storage
│── Dockerfile                         # Docker container setup
│── docker-compose.yml                  # Multi-container setup (API + Pinecone)
│── requirements.txt                    # Dependencies
│── .env.example                        # Environment variables template
│── README.md                           # Project documentation

---

🛠️ Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/Heban-7/rag-chatbot.git cd rag-chatbot

2️⃣ Set Up Environment Variables

Create a .env file (or rename .env.example) and configure:

PINECONE_API_KEY=your_pinecone_api_key
BASE_URI=https://api.openai.com
API_KEY=your_openai_api_key
MODEL_NAME=gpt-4
INDEX_NAME=your_pinecone_index
NAMESPACE=your_pinecone_namespace

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Start the FastAPI Server

uvicorn app:app --host 0.0.0.0 --port 8000

🐳 Running with Docker

docker-compose up -d --build