One hands-on playground for each idea in your handbook, plus a scored quiz on every page drawn from the chapter's own exercises. Tap any card and start playing — it all runs in your browser.
The six stages of building an AI — from defining the problem to monitoring it — plus how a model trains and what makes something AI instead of plain automation.
Walk the 6 stages, sort Automation vs AI, and run a tiny student-score predictor.
Play → 🗃️🗃️Stage 2 in action — clean data, then split it into training, validation & test.
Play → 🤖🤖Watch a model learn from text (stage 3) then answer one token at a time.
Play →AI for the environment, healthcare, automation and education — and the no-code tools (like Teachable Machine) that let you train an AI by just giving examples.
Explore 6 real-world sectors, then play match-the-application.
Play → 👁️👁️How an image AI's first layer "sees" — like the no-code image projects you build.
Play → 🗂️🗂️The same idea as a Teachable-Machine cat/dog model — labelled examples → a boundary.
Play →AI learns only from the data we give it — so balanced data matters. See how unfair data makes an unfair AI, and how the right examples fix it.
Unbalance the 🏏/🏸 training data and watch accuracy skew — then balance it back.
Play → 🗂️🗂️See how the examples you show a model decide what it learns.
Play →