The 14-week curriculum
Every week opens on its first day. Your place is remembered in this browser, so you can stop and come back.
Start here
Pick any week. Week 1 assumes no prior experience.
Foundations
What AI is, how machines learn, and the methods that came beforeWhat Artificial Intelligence Actually Is
Rules, search and learning taken apart and compared on the same problem, with the habits that make any result trustworthy.
7 daysStart day 1 → Week 02Not startedMeasuring a Model Honestly
Why accuracy misleads, what to use instead, and the four ways a good result turns out not to be one.
7 daysStart day 1 → Week 03Not startedSearch, Knowledge and Reasoning
The half of AI that derives answers instead of learning them, and why it is still the right tool for a great deal of real work.
7 daysStart day 1 →Language
Text as data, the standard tasks, embeddings and transformersText as Data
How language becomes numbers, and why that step decides more than the choice of model.
7 daysStart day 1 → Week 05Not startedThe Standard Language Tasks
Entity extraction, keywords, topics, similarity and sentiment, each with the measurement it actually needs.
7 daysStart day 1 → Week 06Not startedMeaning and Embeddings
How a machine gets from words as arbitrary symbols to words with relationships, built from counting and compression.
7 daysStart day 1 → Week 07Not startedTransformers and Large Language Models
Contextual representations, attention worked through by hand, and an honest comparison against the simple methods.
7 daysStart day 1 →Vision
Images as data, classification, and what comes after itImages as Data
What a picture looks like to a computer, and the classical methods for turning it into something usable.
7 daysStart day 1 → Week 09Not startedClassifying Images
Six classifiers compared honestly on the same images, including the modern one that does not win.
7 daysStart day 1 → Week 10Not startedBeyond Classification
Detection, segmentation and transfer learning, and the failures that test accuracy cannot show you.
7 daysStart day 1 →Responsibility
Bias, explainability, privacy and governanceBias and Fairness
Measured on data whose truth is known, including the definitions of fairness that cannot all be satisfied at once.
7 daysStart day 1 → Week 12Not startedExplaining What a Model Did
Importance, counterfactuals and partial dependence, with the traps that make each of them misleading.
7 daysStart day 1 → Week 13Not startedPrivacy, Safety and Governance
Re-identification, what a model leaks about its training data, and the documents a deployed system needs.
7 daysStart day 1 →Capstone
Your own project, and where to go nextEvery week ends with a self-assessment quiz, and each day can be marked complete as you go. Nothing is sent anywhere; your progress lives in this browser only.
Programme detailsAsk a question