AI, Unplugged: Local LLMs

Almost everything people build with language models runs through somebody else's API. This workshop was about the other option: getting a model running locally on your own laptop.

It's worth doing at least once. Running a model yourself makes the tradeoffs concrete — what fits in memory, what speed you actually get, what quantisation costs you — in a way that reading about it doesn't. It also means your data never leaves the machine.

Hands-on throughout, on whatever laptop you brought.