welcome

To get the most from it, you should be comfortable with Python and the command line. You do not need any prior experience with fine-tuning, machine learning theory, or GPUs; we teach the LLM-specific parts from the ground up. Every hands-on example runs on a single GPU you can actually get hold of, and much of it on a laptop. We use a small open-weights model on purpose, so you can run everything yourself and learn by doing. Enterprises mostly run much larger models on far more compute, and that is the point: the techniques here are model-size-agnostic, so what you learn on the small model applies unchanged at that scale, where the difference is more memory and compute, not different methods.

We wrote this book because we kept answering the same question: a team has a real use case and a hardware budget, and asks which adaptation technique to use, and how. A general-purpose model API is a fine start. Still, most teams hit a wall: cost at scale, data leaving boundaries it should not cross, latency, or a model that does not know their domain, house terminology, or internal tools. The fix is to adapt a model you control, and the answers to "how" were scattered across papers and disconnected tutorials. We wanted one reproducible, honest reference instead.