This repository provides a structured curriculum for developers, security professionals, and operations engineers to learn about Large Language Models (LLMs) and Generative AI. It bridges the gap between traditional software engineering and the rapidly evolving AI landscape, offering practical lessons focused on development, operations, and security aspects of LLM integration.

How It Works

The lessons are primarily built around the Langchain framework, guiding users through core LLM concepts and practical applications. The approach emphasizes a narrative structure, making complex topics accessible to those with a Python background. Key areas covered include prompt engineering, embeddings, vector databases, Retrieval Augmented Generation (RAG), agent implementation, cost management, caching strategies, local LLM deployment, and security considerations like prompt injection and output validation.

Quick Start & Requirements

• Installation: Recommended via DevContainer for local setup. Alternatively, Google Colab notebooks are available.
• Prerequisites: Python 3.11+ (examples tested with 3.12), Poetry for virtual environment management, VSCode with Python and Jupyter extensions, and ipykernel.
• Setup: Detailed instructions for configuring VSCode with Poetry and Python interpreters are provided.
• Resources: Links to lessons and the repository structure are available within the README.

Highlighted Details

• Covers LLM concepts from Development, Security, and Operations perspectives.
• Utilizes the Langchain framework for practical examples and RAG implementation.
• Includes lessons on local LLM deployment with Ollama and security threats like prompt injection.
• Offers guidance on cost management, caching, and output structuring.

Maintenance & Community

The repo is actively maintained, with lessons refined for meetups and hackathons. Contributions are welcomed via GitHub issues for new topics or corrections.

Licensing & Compatibility

The repository does not explicitly state a license in the provided README. Users should verify licensing for commercial or closed-source integration.

Limitations & Caveats

The README notes that the "Requirements to run this repo (needs more love)" section indicates potential areas for improvement in setup documentation or tooling. Some examples may have syntax specific to older Langchain versions, though updates are noted.