Introduction

​

What is LangChat?

​

Why LangChat?

• Integrate LLM APIs (OpenAI, Anthropic, etc.)
• Manage Vector Databases (Pinecone, Weaviate, etc.)
• Handle Chat History (conversation context and memory)
• Implement Reranking (improve search result relevance)
• Track Metrics (response times, errors, feedback)
• Rotate API Keys (handle rate limits and failures)

• 🚀 Fast Development: Get a production-ready chatbot running in minutes
• 🔧 Modular Architecture: Use components independently or together
• 🔄 Automatic API Key Rotation: Built-in fault tolerance for OpenAI API
• 📊 Production Ready: Includes chat history, metrics, and feedback tracking
• 🎨 Highly Customizable: Easy to extend with custom prompts and adapters
• 🐳 Docker Ready: Auto-generated Dockerfile for easy deployment

​

Key Features

​

🤖 LLM Integration

• OpenAI: Native OpenAI API support with automatic API key rotation
• Fault Tolerant: Automatic retry logic with multiple API keys
• Production Ready: Handles rate limits and errors gracefully

​

🔍 Vector Search

• Pinecone Integration: Seamless vector database integration
• Reranking: Flashrank reranker for improved search results
• Configurable Retrieval: Adjustable document retrieval and reranking

​

💾 Database Management

• Supabase: Built-in Supabase integration
• ID Management: Automatic ID generation with conflict resolution
• Session Management: User-specific chat history and memory

​

🎨 Customization

• Custom Prompts: Configure both system prompts and standalone question prompts
• Flexible Configuration: Environment variables or direct configuration
• Modular Architecture: Use components independently or together

​

🚀 Developer Experience

• Auto-Generated Interface: Chat interface HTML auto-created on startup
• Auto-Generated Dockerfile: Dockerfile auto-created with correct port
• Easy Setup: Simple configuration and initialization

​

Quick Example

import asyncio
from langchat import LangChat, LangChatConfig

async def main():
    # Create configuration
    config = LangChatConfig(
        openai_api_keys=["your-openai-api-key"],
        pinecone_api_key="your-pinecone-api-key",
        pinecone_index_name="your-index-name",
        supabase_url="your-supabase-url",
        supabase_key="your-supabase-key"
    )
    
    # Initialize LangChat
    langchat = LangChat(config=config)
    
    # Chat with the AI
    result = await langchat.chat(
        query="What are the best universities in Europe?",
        user_id="user123",
        domain="education"
    )

asyncio.run(main())

​

Architecture Overview

┌──────────────────────────────────────────────┐
│                LangChat Engine                                           │
│     (Main orchestrator for all components)                               │
└──────────────────────────────────────────────┘
                      │
                      ├─── LLM Service (OpenAI)
                      │       └─── Automatic key rotation
                      │
                      ├─── Vector Database (Pinecone)
                      │       └─── Document retrieval & embeddings
                      │
                      ├─── Reranker (Flashrank)
                      │       └─── Result relevance improvement
                      │
                      ├─── Database (Supabase)
                      │       ├─── Chat history storage
                      │       ├─── Metrics tracking
                      │       └─── Feedback collection
                      │
                      └─── Session Management
                              └─── User context & memory

​

Use Cases

• 📚 Education Chatbots: Help students find universities and programs
• ✈️ Travel Assistants: Provide travel recommendations and booking guidance
• 🛒 Customer Support: Answer product questions with RAG (Retrieval-Augmented Generation)
• 💼 Business Assistants: Internal knowledge base queries
• 🎓 Learning Platforms: Answer questions about course materials
• 🏥 Healthcare: Provide medical information with context

​

What Makes LangChat Different?

​

Next Steps

1. Getting Started - Set up LangChat in minutes
2. Installation Guide - Install and configure LangChat
3. Configuration - Learn about all configuration options
4. Examples - See LangChat in action

​

Community & Support

• GitHub: github.com/neurobrains/langchat
• Issues: Report bugs or request features
• Contributions: We welcome contributions!