LLMs and a Bit More

A comprehensive educational repository exploring Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), AI Agents, and Model Context Protocol (MCP) through practical, hands-on examples.

What's Inside

This repository contains practical demonstrations and implementations of modern AI technologies:

LLM Demos (llms.ipynb)

• Azure OpenAI Integration: Complete setup guide for Azure OpenAI services
• Chat Conversations: Build an AI study buddy for computer science topics
• Image Generation: Create visual aids using DALL-E
• Educational Applications: Real-world examples for students
• Essay Writing Assistant: Help with academic writing and outlines

RAG Implementation (rag.ipynb)

• Document Processing: Extract text from PDF files (Constitution of Kenya 2010)
• Text Chunking: Intelligent document segmentation
• Vector Embeddings: Using Sentence Transformers for semantic search
• FAISS Integration: Efficient similarity search
• Query System: Ask questions about constitutional content

Spotify MCP Server (spotify-play-music/)

A complete Model Context Protocol (MCP) server for Spotify integration:

• Music Control: Play, pause, skip tracks, and adjust volume
• Apple Script Integration: Native macOS Spotify control
• MCP Tools: Structured tool definitions for AI agents
• Debug Support: Built-in debugging with MCP Inspector
• Agent Builder Integration: Connect with AI Toolkit Agent Builder

Getting Started

Prerequisites

• Python 3.8+
• Azure OpenAI subscription (for LLM demos) or GitHub Models
• Spotify application (for MCP server)
• macOS (for Spotify MCP server)

Setup Instructions

1. Clone the repository:

   git clone https://github.com/BethanyJep/llms-and-a-bit-more.git
   cd llms-and-a-bit-more

2. For LLM Notebooks:

   • Create a .env file with your Azure OpenAI credentials:

     AZURE_OPENAI_ENDPOINT=your_azure_endpoint_here
     AZURE_OPENAI_API_KEY=your_azure_api_key_here

   • Install dependencies:

     pip install openai python-dotenv requests pillow

3. For RAG Implementation:

   pip install pymupdf sentence-transformers faiss-cpu openai tiktoken

4. For Spotify MCP Server:

   cd spotify-play-music
   # Using uv (recommended)
   uv venv && uv pip install -r pyproject.toml --extra dev
   # Or using pip
   python -m venv .venv && pip install -e .[dev]

Spotify MCP system prompt

System ptompt: Control Spotify playback effectively using specific tools for playing, pausing, skipping tracks, setting volume, and retrieving currently playing track information.

Usage Examples

LLM Chat Assistant

# Create a conversation with AI
study_conversation = [
    {"role": "system", "content": "You are a helpful computer science tutor."},
    {"role": "user", "content": "Explain data structures simply."}
]
response = chat_with_ai(study_conversation)

RAG Query System

# Query the Constitution of Kenya
question = "What are the fundamental rights in Kenya?"
relevant_chunks = query_constitution(question, k=3)

Spotify MCP Tools

• play_music(query): Play specific songs, artists, or albums
• pause_music(): Pause current playback
• next_track() / previous_track(): Navigate tracks
• get_current_track(): Get currently playing track info
• set_volume(level): Adjust volume (0-100)

Resources

• GenAI for Beginners - Microsoft's comprehensive guide to Generative AI
• AI Agents for Beginners - Learn about AI agents and autonomous systems
• MCP for Beginners - Model Context Protocol fundamentals
• GitHub for Students - Free tools and resources for students
• GitHub Models - Explore AI models on GitHub
• Azure for Students - Free Azure credits for students

Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes, please open an issue first to discuss what you would like to change.

License

This project is licensed under the MIT License.