LLM Engineering Project

This repository contains a structured learning path for LLM (Large Language Model) engineering, organized into weekly modules that cover various aspects of working with AI models, from API integration to model deployment.

Project Overview

This project provides hands-on experience with:

• OpenAI API integration
• Local LLM deployment with Ollama
• Hugging Face models and transformers
• Text generation, summarization, and classification
• Image generation with stable diffusion models
• Text-to-speech and speech-to-text conversion
• Building interactive AI applications with Gradio

Project Structure

llm_engineering/
├── main.py                # Main entry point
├── week1/                 # Introduction to OpenAI APIs
│   ├── day1.ipynb         # OpenAI API basics and web scraping
│   ├── day2.ipynb         # Content summarization
│   └── day5.ipynb         # Advanced techniques
├── week2/                 # Working with local LLMs
│   ├── day1.ipynb         # Ollama setup
│   ├── day2.ipynb         # Building simple chatbots
│   ├── day5.ipynb         # Function calling and text-to-image generation
├── week3/                 # Advanced models & multimodality
│   ├── day1.ipynb         # Hugging Face transformers
│   ├── day2.ipynb         # Multimodal applications
│   ├── day3.ipynb         # Advanced transformers
│   ├── day4.ipynb         # Quantized models
│   ├── day5.ipynb         # Meeting minutes generator
│   ├── speech.wav         # Sample audio file
│   └── denver_extract.mp3 # Meeting audio sample
└── .env                   # Environment variables for API keys

Functionality Summary

Week 1: Getting Started with LLMs

• Website content scraping and analysis
• OpenAI API integration for text summarization
• Prompt engineering techniques
• Structured data extraction

Week 2: Local Models & Interactive Applications

• Local LLM setup with Ollama
• Building an airline assistant chatbot
• Function calling with LLMs for dynamic responses
• Text-to-image generation with Stable Diffusion
• Creating interactive interfaces with Gradio

Week 3: Advanced Multimodal Applications

• Hugging Face pipeline integrations (sentiment analysis, NER, QA)
• Text-to-image generation with multiple models
• Speech synthesis and audio processing
• Building multimodal applications
• Quantized LLM inference with 4-bit precision
• Automated meeting minutes generation from audio recordings
• Speech-to-text with Whisper models

Requirements

This project requires Python 3.13 or higher with the dependencies listed in the pyproject.toml file, including:

• transformers - For accessing Hugging Face models
• diffusers - For text-to-image generation
• torch - PyTorch for tensor operations
• faster-whisper - For efficient speech recognition
• bitsandbytes - For model quantization
• requests/httpx - For API communications

Setup Instructions

1. Clone the repository:

   git clone <repository-url>
   

2. Create a virtual environment:

   python -m venv .venv
   

3. Activate the virtual environment:

   • Windows: .venv\Scripts\activate
   • Linux/Mac: source .venv/bin/activate

4. Install dependencies:

   pip install -e .
   

5. Set up environment variables:

   • Create a .env file based on the provided .env.example
   • Add your API keys for OpenAI and Hugging Face

Running the Project

• Individual notebooks can be run in Jupyter:

  jupyter notebook
  

License

This project is provided for educational purposes.