A hands-on LLM course that admits it’s still under construction

What it does

This repository hosts a free, hands-on LLM engineering course built from Jupyter notebooks and companion Medium articles. It covers the OpenAI API, Hugging Face, LangChain, vector databases, and fine-tuning techniques like LoRA and QLoRA, organized into three sections: techniques and libraries, projects, and enterprise solutions. The material doubles as the unofficial, continually updated companion to an Apress book—though the README notes the GitHub course does not contain everything in the printed version.

The interesting bit

The author is admirably upfront that the course is “in permanent development,” and the notebooks are explicitly designed to outlive the book by incorporating new examples and chapters. There is something refreshing about educational content that admits some articles are still WIP and that you will probably need more memory than free Colab provides.

Key highlights

• Covers practical integration of OpenAI, Hugging Face, LangChain, ChromaDB, and Faiss
• Includes specific small projects like a restaurant chatbot, NL2SQL translator, and semantic cache for RAG
• Each lesson pairs a runnable notebook with a Medium article for deeper explanation
• Addresses memory constraints honestly, noting that some notebooks exceed free Colab limits
• Topics span from prompt engineering and PEFT to knowledge distillation and LLMOps

Caveats

• The course is incomplete by design; some lessons and articles are marked WIP
• Several notebooks require more memory than free-tier Colab provides, so expect to run locally or pay for Pro
• The relationship between the repo and the Apress book is slightly confusing; the README calls this the “unofficial repository” yet states the book is based on this content

Verdict

Worth bookmarking if you are an engineer or researcher who learns best by running code and reading alongside it. Skip it if you need a polished, end-to-end certification or guaranteed completeness.