README.md

FastAPI RAG Chatbot Backend

This project implements a FastAPI backend for a Retrieval-Augmented Generation (RAG) chatbot. It integrates with OpenAI for chat completions and embeddings, Neon Serverless Postgres for persistent data, and Qdrant Cloud for vector storage.

Project Structure

backend/
├── app/
│   ├── __init__.py
│   ├── main.py
│   ├── config.py
│   ├── database.py
│   ├── qdrant_client.py
│   ├── models/
│   │   ├── __init__.py
│   │   ├── user.py
│   │   └── chat.py
│   ├── schemas/
│   │   ├── __init__.py
│   │   └── chat.py
│   ├── routes/
│   │   ├── __init__.py
│   │   ├── chat.py
│   │   └── health.py
│   └── services/
│       ├── __init__.py
│       ├── rag_service.py
│       ├── embeddings_service.py
│       └── openai_service.py
├── scripts/
│   └── ingest_content.py
├── .env.example
├── requirements.txt
├── README.md
└── .gitignore

Setup Instructions

1. Clone the repository:

   git clone <repository-url>
   cd robotic
   

2. Navigate to the backend directory:

   cd backend
   

3. Set up the Python virtual environment and install dependencies:

   On Windows, run:

   .\setup.bat
   

   On Linux/macOS, run:

   python3 -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt
   

4. Configure Environment Variables:

   Create a .env file in the backend/ directory by copying .env.example and filling in your credentials:

   copy .env.example .env
   # or for Linux/macOS
   cp .env.example .env
   

   Edit the .env file with your actual API keys and database URLs:

   OPENAI_API_KEY=your_openai_api_key_here
   NEON_DATABASE_URL=your_neon_postgres_connection_string_here
   QDRANT_URL=your_qdrant_cluster_url_here
   QDRANT_API_KEY=your_qdrant_api_key_here
   

Running the Application

1. Activate your virtual environment:

   On Windows:

   .\venv\Scripts\activate
   

   On Linux/macOS:

   source venv/bin/activate
   

2. Start the FastAPI server:

   On Windows, run:

   .\run.bat
   

   On Linux/macOS, run:

   uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
   

   The API will be accessible at http://localhost:8000.

API Endpoints

(Note: Detailed API documentation will be available at /docs once the server is running)

Health Check

• GET /health
  • Returns the health status of the backend and its integrated services.

Chatbot Interaction

• POST /chat/

  • Description: Sends a user query to the RAG chatbot and receives a generated response.

  • Request Body Example:

    {
        "query": "What is the main topic of the book?"
    }
    

  • Response Body Example:

    {
        "response": "The main topic of the book is ..."
    }
    

Content Ingestion Script

• scripts/ingest_content.py

  • This script is responsible for reading MDX files from ../physical-ai-humanoid-robotics/docs/, chunking the text, generating OpenAI embeddings, and storing them in Qdrant.

  • Usage:

    python scripts/ingest_content.py