A free, structured AI course in three tracks: foundations, practical prompting and real working AI workflows. Written for people who use AI at work, not researchers.
Each lesson is self-contained and takes six to eight minutes. Work through a track in order if you are new, or jump straight to the lesson that matches the problem in front of you. Every lesson links to the tools you can practise with and to the glossary for any term you have not met.
What these models actually are, how they fail, and how to judge output like a professional rather than a spectator.
A working mental model of language models — tokens, context, prediction — that explains almost every behaviour you will encounter.
A fast, repeatable review process that catches hallucinations, filler and structural weakness before anything gets published.
The practical rules for using AI at work without creating a legal, ethical or reputational problem.
The repeatable patterns that separate a usable first draft from three rounds of rewriting.
Five components that turn a vague request into a specification the model can actually satisfy.
Structured techniques for improving output: single-variable iteration, critique loops, and decomposition.
Subject, composition, lighting and style — the four levers that decide whether a generated image is usable.
Turning one-off prompts into reliable systems for writing, code, research and marketing.
An end-to-end process for producing publishable long-form writing without the flat, machine-made texture.
Using models for real engineering work: scaffolding, debugging, review and tests, with a clear line around what to never trust.
How to use AI on documents, data and decisions while keeping every claim traceable to a real source.