Advanced RAG-Powered Longevity Coach

[work in progress]

A secure RAG (Retrieval-Augmented Generation) application that delivers personalized health and longevity advice. Built with LangChain and Streamlit, it combines hybrid search capabilities with user-controlled data management and Google OAuth authentication.

🎯 Key Features

🔐 Secure Authentication System

• Google OAuth2 Integration: Secure login with Google accounts
• Session Management: Automatic session timeout and secure logout
• Protected Routes: All pages require authentication for access
• User Context: Maintains user identity throughout the application

🧠 Advanced RAG System

• Hybrid Search: Combines BM25 (keyword) and FAISS (semantic) search for optimal relevance
• LangChain Integration: Built on LangChain for extensible retrieval and processing
• Multi-Strategy Retrieval: Contextual compression and rank fusion for enhanced results
• Real-Time Processing: Live feedback during search planning and document processing

📊 Flexible Knowledge Management

• Interactive Data Editor: Spreadsheet-like interface for direct knowledge base editing
• PDF Document Processing: Automatic extraction and structuring of uploaded documents
• Guided Conversational Entry: AI-assisted data entry through natural language chat
• Automatic Re-indexing: Seamless updates to search indices when data changes

🔧 Modern Architecture

• Type-Safe Design: Comprehensive Pydantic models and type hints throughout
• Modular Components: Clear separation of concerns with extensible provider system
• Robust Error Handling: Custom exception hierarchy with informative error messages
• Configurable Settings: Environment-driven configuration with sensible defaults

🚀 Getting Started

Prerequisites

• Python 3.9+
• Google Cloud Project (for OAuth2 authentication)
• OpenAI API key (for LLM functionality)

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/LLM-RAG-Longevity-Coach.git
   cd LLM-RAG-Longevity-Coach

2. Install dependencies:

   pip install -r requirements.txt

3. Set up environment variables:

   cp .env.example .env

   Edit .env with your configuration:

   # Required: OpenAI API key
   OPENAI_API_KEY=your_openai_api_key_here
   
   # Required: Google OAuth2 credentials
   GOOGLE_CLIENT_ID=your_google_client_id_here
   GOOGLE_CLIENT_SECRET=your_google_client_secret_here
   OAUTH_REDIRECT_URI=http://localhost:8501/
   
   # Optional: Development settings
   OAUTH_INSECURE_TRANSPORT=true  # For local development

4. Set up Google OAuth2:

   • Go to the Google Cloud Console
   • Create a new project or select an existing one
   • Enable the Google+ API
   • Create OAuth2 credentials (Web application)
   • Add http://localhost:8501/ to authorized redirect URIs
   • Copy the client ID and secret to your .env file

5. Run the application:

   streamlit run app.py

First Time Setup

1. Navigate to http://localhost:8501
2. Click "Login with Google" to authenticate
3. Once logged in, you'll have access to all features with your own private knowledge base

📋 How It Works

The application provides two main workflows after authentication:

🗣️ Chat Interface (Main Page)

1. Authentication Check: Secure login required for access
2. Query Analysis: AI analyzes your health/longevity question
3. Search Strategy: System determines optimal search approach
4. Hybrid Retrieval: Combines BM25 and FAISS search for maximum relevance
5. Response Generation: LLM generates personalized advice using retrieved context
6. Real-Time Feedback: Live updates showing search and processing steps

📚 Knowledge Base Management

Choose from multiple data input methods:

Interactive Data Editor (Knowledge Base page)

• Spreadsheet-like interface for direct editing
• Add, modify, or delete entries in real-time
• Automatic data validation and error handling
• One-click re-indexing of the search system

PDF Document Upload (Upload Documents page)

• Drag-and-drop PDF processing
• LLM-powered text extraction and structuring
• Automatic conversion to structured knowledge entries
• Progress tracking and error handling

Guided Conversational Entry (Guided Entry page)

• Natural language data entry through chat interface
• AI assistant helps structure your information
• Review and approve entries before saving
• Perfect for adding personal health insights

🔧 Configuration Options

The application supports extensive configuration through environment variables:

Core Settings

# LLM Configuration
DEFAULT_LLM_MODEL=o3                    # OpenAI model to use
DEFAULT_TEMPERATURE=1.0                 # Response creativity (0.0-2.0)
GOOGLE_API_KEY=your_key                 # Optional: for Gemini models

# Search Configuration
DEFAULT_TOP_K=5                         # Number of documents to retrieve
VECTOR_STORE_FOLDER=vector_store_data   # Local storage location
EMBEDDING_MODEL=text-embedding-3-large  # OpenAI embedding model

# Authentication
SESSION_TIMEOUT_HOURS=24                # Session timeout duration
OAUTH_INSECURE_TRANSPORT=true          # Enable for local development

Advanced Options

# Performance Tuning
RRF_K=60                               # Rank fusion parameter
SEARCH_MULTIPLIER=2                    # Search result multiplier
MAX_DOCUMENT_LENGTH=10000              # Document chunk size limit

# Data Management
DOCS_FILE=docs.jsonl                   # Knowledge base file name
MAX_INSIGHTS=5                         # Maximum insights per query
MAX_CLARIFYING_QUESTIONS=3             # Maximum clarifying questions

🛡️ Security & Privacy

• OAuth2 Security: Industry-standard authentication with Google
• Session Management: Automatic timeout and secure logout
• Local Data Storage: All user data stored locally (no cloud dependency)
• API Key Protection: Secure handling of OpenAI API keys
• Input Validation: Comprehensive validation of all user inputs
• Protected Access: All application features require authentication

Note: This application currently uses a shared knowledge base for all authenticated users. Multi-tenant data isolation is planned for a future release.

🏗️ Architecture

Built with modern Python technologies:

• Frontend: Streamlit for interactive web interface
• Backend: LangChain for RAG pipeline and document processing
• Search: FAISS for vector similarity + BM25 for keyword matching
• Authentication: Google OAuth2 with secure session management
• Data Models: Pydantic for type-safe data validation
• Storage: Local JSONL files with automatic indexing

🚨 Troubleshooting

Common Issues

Authentication Problems:

• Ensure Google OAuth2 credentials are correctly configured
• Check that redirect URI matches exactly (including trailing slash)
• Verify Google+ API is enabled in your GCP project

Installation Issues:

• Use Python 3.9+ for compatibility
• Install dependencies in a virtual environment
• Check that all required environment variables are set

Performance Issues:

• Increase DEFAULT_TOP_K for more comprehensive search
• Adjust SEARCH_MULTIPLIER for broader result sets
• Consider using a more powerful OpenAI model

Data Management:

• Ensure DOCS_FILE location is writable
• Check file permissions for VECTOR_STORE_FOLDER
• Verify PDF files are not corrupted during upload

Getting Help

• Check the Issues page for known problems
• Review the CLAUDE.md file for detailed technical documentation
• Examine application logs for specific error messages

🤝 Contributing

Contributions are welcome! Please feel free to submit pull requests, report bugs, or suggest new features.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📜 License

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

🙏 Acknowledgments

• Built with LangChain for RAG capabilities
• Streamlit for the web interface
• OpenAI for LLM and embedding models
• Google Cloud for OAuth2 authentication
• FAISS for vector similarity search

---

Ready to get started with your personalized longevity coach? Follow the installation steps above and begin building your own health knowledge base today!