LangChain Learning Project

A clean, modular LangChain project designed for learning and experimentation with multiple LLM providers.

Project Overview

This project provides a flexible framework for working with different Language Model providers (OpenAI, Anthropic, OpenRouter) through a unified interface. It's built with Docker for easy deployment and Jupyter notebooks for interactive learning.

Quick Start

# Clone and navigate to project
cd langechain

# Start all services
docker compose up --build

# Access services
- API: http://localhost:8000
- Jupyter: http://localhost:8888
- Vector DB: http://localhost:8001

Project Structure

langechain/
├── config/          # Configuration files (YAML)
├── data/            # Data storage and cache
├── examples/        # Example usage scripts
├── notebooks/       # Jupyter notebooks for learning
├── src/             # Main source code
├── tests/           # Unit and integration tests
└── docker-compose.yaml  # Container orchestration

Key Features

• 🔄 Multi-Provider Support: OpenAI, Anthropic, OpenRouter
• 🎯 Model Auto-Detection: Automatic provider routing by model name
• 📝 Model Aliases: Use simple names like gpt4, claude, mistral
• 🔧 Configuration-Driven: YAML-based setup
• 🐳 Docker Ready: Containerized development environment
• 📚 Learning Focused: Jupyter notebooks and examples
• 🔗 Session Management: Conversation history support
• 🌐 REST API: FastAPI-based endpoints

Environment Setup

Copy .env.example to .env and add your API keys:

OPENAI_API_KEY="your-openai-key"
ANTHROPIC_API_KEY="your-anthropic-key"
OPENROUTER_API_KEY="your-openrouter-key"

Documentation

• See individual folder README files for detailed information
• Check notebooks/ for interactive tutorials
• Review examples/ for usage patterns

Learning Path

1. Start with: notebooks/expiriment.ipynb
2. Explore: examples/chat_session.py
3. Configure: config/model_config.yaml
4. Build: Create your own LLM applications

Contributing

This is a learning project. Feel free to experiment, break things, and learn from the process!

License

MIT License - See LICENSE file for details