2026-10-26.2 · 40 minutes estimated; work in several sessions
Saved on this device; sign in to import explicitly.
Prerequisites and goal
A real experience card or explainable work sample.
Start with diagnostics; skip explanations you can already apply independently.
Organize around evidence
Answer directly, then give necessary context, actions, results and reflection. STAR is a structure, not equal-length sections. Explain decisions in behavioral answers, and assumptions and checks in technical ones. State limits and ask clarifying questions when needed.
Ask AI to probe weak evidence
Record a first answer. Have AI quote and probe contribution, alternatives and failure conditions. After revision, answer a different scenario independently. Distinguish knowledge gaps, unsuitable examples, verbosity and anxiety rather than assigning one generic score.
Worked example · fictional
From a problem to inspectable work
Fictional: a leadership answer spends 90 seconds on background and one sentence on action.
Shorten context; retain the conflict and personal decision.
Add a follow-up about the rejected alternative.
Use most of two minutes for task allocation, delays and checks; keep outcomes separate from attribution.
Failure case and repair
Stop practising after AI gives a perfect score.
A score does not establish transfer to unfamiliar follow-up.