Interactive guide
Make the Decision Without Faking Certainty
A subtle decision-system failure can return an answer that looks plausible even when its evidence or meaning is wrong.
For example, a missing field might be treated as a bad signal rather than an unavailable input. A score of 20 might be read as “a bit unhealthy” even if most of its points come from email activity. A confidence figure of 0.9 might be mistaken for a nine-in-ten chance of being right, though the number may not have been calibrated against outcomes. Each case needs its own evidence and limitations made visible.
You will build a Decision Design: a stated decision shape, explicit rules, a bounded approach to interpretation, clear handling for missing and conflicting evidence, an honest confidence policy, a consistent output shape, and a set of awkward cases to test against a later implementation. The lab saves your progress in this browser.
Work through the evidence with Maya.
Compare rules, scores, and interpretation in the prepared cases. Build a decision design that explains its result and makes missing or conflicting evidence visible.
Start with a broken answer
Pick the Right Decision Shape
Chapter 1 of 8 · 16 min
The learning path
Every chapter starts from a plausible wrong answer and works backwards to whatever produced it. An empty field read as a bad signal. A score of 20 that means two unrelated things. A confidence figure that measures something other than what the person reading it assumes.
Choosing how the answer gets made
- 01Pick the Right Decision ShapeChoose between classifying, scoring, ranking, and recommending, and know what each can and cannot answer.~16 min · exercise
- 02Apply Simple Rules FirstDo the deterministic work first, and let one condition short-circuit the rest.~14 min · exercise
- 03Use AI Where Interpretation HelpsDraw the line between what a rule can check and what needs reading, and put the model on one side of it.~16 min · exercise
- 04Handle Missing and Conflicting EvidenceStop absence being read as a bad signal, and decide which source wins when two disagree.~16 min · exercise
Surviving contact with reality
- 05Treat Confidence CarefullySay what a confidence number does and does not mean, and keep a composite score’s parts recoverable.~14 min · exercise
- 06Return a Consistent ResultReturn the same shape every time, carrying what the next guide needs to explain it.~14 min · exercise
- 07Test Plausible Edge CasesBuild cases from documented failures, near misses, and boundaries; adapt the rules when populations require different evidence.~18 min · exercise
- 08Capstone: Validate the Decision DesignReview prepared examples, validate the saved design, and approve only its current version.~20 min · exercise