Your goal: hold one level across two rounds, the way a loop is decided.
How this works (about 45 minutes)
- Round 1, risk (about 18 minutes): name and rank the risks of an AI deployment, with follow-ups and a curveball.
- Round 2, build and cost (about 27 minutes): design an agent with retrieval, read a pilot sheet Felix holds, handle a curveball and make a cost call. Ask him for facts; they are made up for the exercise.
- Felix tells you when he switches rounds. This is a practice room in the style of an AI lab loop, built from what candidates report, not the company's real process.
- If you want to sketch, ask Felix for the notepad and tell him when you have written something.
- No hints and no scores during the call. Your report comes after.
What scores well
- Risks named across misuse, errors, bias, privacy, security and over-reliance, then ranked with a reason
- Mitigations in layers with a way to test them, and what you cannot prove
- A design that budgets tokens and keeps a person on the decision
- A sheet read by slice and by cost per correct decision
- Calibrated confidence: what you know, what you do not, and how you would find out
Tips
- Take the Gen AI coaches for evaluation, security, responsible AI and agents first.
- Say plainly what you are unsure about.