NotebookLM Clone - Multi-Agent AI Application

A full-stack AI application similar to Google NotebookLM, built with a multi-agent architecture for document ingestion, semantic search, and LLM-powered question-answering with citations.

🏗️ Architecture

Multi-Agent System

1. Parser Agent - Handles document processing (PDF, DOCX, TXT, URLs)
2. Embedding Agent - Converts text chunks to vector embeddings
3. Retriever Agent - Performs semantic search over vector database
4. RAG Prompting Agent - Generates answers with LLM integration and citations
5. Chat History Agent - Manages conversation sessions and context

Tech Stack

Backend (FastAPI + Python)

• FastAPI for REST API
• SQLAlchemy for database ORM
• ChromaDB for vector storage
• Sentence Transformers for embeddings
• OpenAI/Anthropic API integration
• JWT authentication

Frontend (React + TypeScript)

• React 18 with TypeScript
• Tailwind CSS for styling
• React Query for state management
• Zustand for auth state
• React Router for navigation

Database & Storage

• SQLite/PostgreSQL for metadata
• ChromaDB for vector embeddings
• Local file storage for documents

🚀 Features

• Document Upload: Support for PDF, DOCX, TXT files and web URLs
• Semantic Search: AI-powered search across all documents
• Question Answering: Get answers with source citations
• Chat Sessions: Threaded conversations with context
• Document Management: Upload, view, and organize documents
• User Authentication: Secure login and user management

📋 Prerequisites

• Python 3.8+
• Node.js 16+
• npm or yarn

🛠️ Installation

1. Clone the repository

git clone <repository-url>
cd notebooklm-clone

2. Install dependencies

# Install root dependencies
npm install

# Install all dependencies (frontend + backend)
npm run install:all

3. Environment Setup

Create environment files:

# Backend environment
cp backend/.env.example backend/.env

Edit backend/.env with your configuration:

# Required for LLM integration
OPENAI_API_KEY=your_openai_api_key_here

# Optional - for production
DATABASE_URL=postgresql://username:password@localhost:5432/notebooklm_db
SECRET_KEY=your_secret_key_here

4. Development Setup

Option 1: Run both frontend and backend together

npm run dev

Option 2: Run separately

# Terminal 1 - Backend
cd backend
python dev.py

# Terminal 2 - Frontend  
cd frontend
npm run dev

📚 API Documentation

Once the backend is running, visit:

• API Documentation: http://localhost:8000/docs
• Alternative docs: http://localhost:8000/redoc

🎯 Usage

1. Register/Login: Create an account or sign in
2. Upload Documents: Add PDF, DOCX, TXT files or web URLs
3. Ask Questions: Use the chat interface to query your documents
4. View Citations: See source references for all answers
5. Search: Use semantic search to find relevant content
6. Manage: Organize documents and chat sessions

🔧 Configuration

Backend Configuration

Key environment variables in backend/.env:

# Database
DATABASE_URL=sqlite:///./notebooklm.db

# API Keys
OPENAI_API_KEY=your_key_here
ANTHROPIC_API_KEY=your_key_here

# Security
SECRET_KEY=your_secret_key
JWT_EXPIRE_MINUTES=30

# Storage
CHROMA_PERSIST_DIRECTORY=./chroma_db
UPLOAD_DIR=./uploads
MAX_FILE_SIZE_MB=50

# CORS
ALLOWED_ORIGINS=http://localhost:3000

Frontend Configuration

The frontend automatically proxies API requests to the backend during development.

🧪 Development

Backend Development

cd backend

# Install dependencies
pip install -r requirements.txt

# Run with auto-reload
python dev.py

# Run tests (when implemented)
pytest

Frontend Development

cd frontend

# Install dependencies
npm install

# Start development server
npm run dev

# Build for production
npm run build

# Run linting
npm run lint

📁 Project Structure

notebooklm-clone/
├── backend/
│   ├── app/
│   │   ├── agents/          # Multi-agent system
│   │   ├── models/          # Database models
│   │   ├── routers/         # API routes
│   │   ├── services/        # Business logic
│   │   └── utils/           # Utilities
│   ├── requirements.txt
│   └── dev.py
├── frontend/
│   ├── src/
│   │   ├── components/      # React components
│   │   ├── pages/           # Page components
│   │   ├── services/        # API services
│   │   ├── stores/          # State management
│   │   └── types/           # TypeScript types
│   └── package.json
└── package.json

🚀 Deployment

Backend Deployment

1. Set production environment variables
2. Use a production WSGI server like Gunicorn:

gunicorn app.main:app -w 4 -k uvicorn.workers.UvicornWorker

Frontend Deployment

1. Build the frontend:

cd frontend && npm run build

2. Serve the dist folder with a web server

Docker Deployment (Optional)

Docker configurations can be added for containerized deployment.

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

📝 License

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

🔮 Roadmap

• Real-time collaborative Q&A sessions
• Advanced document processing (images, tables)
• Integration with more LLM providers
• Document version control
• Advanced analytics and insights
• Mobile app support
• API rate limiting and quotas
• Advanced search filters
• Document annotations and highlights

🐛 Known Issues

• Large file uploads may timeout (increase server timeout)
• Vector database rebuilds on restart with SQLite
• Limited file type support (expanding)

📞 Support

For questions or issues:

1. Check the documentation
2. Search existing issues
3. Create a new issue with details

---

Built with ❤️ using modern AI and web technologies