A collection of Jupyter notebooks demonstrating implementations of Large Language Models (LLMs) using LangChain and LangGraph.

This repository contains four notebooks that explore different aspects of LLM application development:

1. Agent.ipynb: Implements a basic agent that can perform web searches using the Tavily API and maintain conversation state.
2. Chatbot.ipynb: Demonstrates building a chatbot with conversation memory and custom prompt templates.
3. Simple LLM Application.ipynb: Covers fundamental LLM interactions and streaming responses.
4. Semantic Search Engine.ipynb: Implements a semantic search system using vector embeddings and document retrieval, with PDF processing and advanced search capabilities.

• LangChain: LLM interactions and prompt management
• LangGraph: State management and agent orchestration
• OpenAI API: Model inference
• Tavily API: Web search functionality
• LangSmith: Observability and tracing
• Python 3.11

• Web search integration using Tavily API
• Basic conversation state management
• Simple memory persistence
• Response streaming

• Conversation memory using LangGraph's MemorySaver
• Custom prompt templates
• Basic token management
• Async processing

• Basic model interactions
• Streaming responses
• Prompt templates
• Environment configuration

• LangSmith integration for tracing and monitoring
• Conversation history tracking
• Performance metrics

• PDF document processing and chunking
• Vector embeddings using OpenAI's text-embedding-3-large model
• In-memory vector store implementation
• Multiple search methods (basic, scored, vector-based)
• Batch processing and async search capabilities
• Document metadata preservation

1. Clone the repository:

   git clone https://github.com/yourusername/llm-playground.git
   cd llm-playground

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Set up environment variables:

   cp .env.example .env

   Then edit .env with your API keys:

   OPENAI_API_KEY=your_openai_key
   TAVILY_API_KEY=your_tavily_key
   LANGSMITH_API_KEY=your_langsmith_key
   LANGSMITH_PROJECT=your_project_name
   

The notebooks demonstrate:

• Basic LLM application development
• API integration (OpenAI, Tavily)
• Simple state management
• Basic async programming
• Environment configuration
• API key management
• Observability with LangSmith

Contributions are welcome. Please submit a Pull Request.

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

For questions or opportunities, please reach out through GitHub or LinkedIn.