14Design an AI copilot for loan officers at a lender. Now compliance says the AI cannot influence the credit decision. What changes?▼hardNewJPMorgan ChaseRevolutKlarna◆ premiumA design question with a regulatory constraint arriving halfway. The interviewer wants a useful product on both sides of it, and a clear line between preparing a decision and making one.Practise this question →Read answer →
09Design an AI writing assistant for our B2B product. Now legal says EU customers' data must never leave the EU. What changes?▼hardNewAtlassianNotionCanva○ sign inA design question with a constraint dropped halfway. The interviewer wants to see residency flow through every component: model calls, logs, caches, embeddings, evals and support access.Practise this question →Read answer →
08We sell an AI add-on at $20 per user per month, flat. Is it profitable? Now the team wants a model that costs three times as much.▼hardNewNotionCanvaAtlassian○ sign inA margin question with a constraint added halfway through. The average user is not the problem; the distribution is, and a flat price hides it until the model gets more expensive.Practise this question →Read answer →
15Every request sends the same 20,000-token system prompt. What would prompt caching save us, and how could it make things worse?▼hardNewAnthropicGoogleCursor◆ premiumCaching is the cheapest large saving in most AI products, and a single line of personalisation at the top of the prompt can turn it into a 22% cost increase. The arithmetic is short; the layout insight is what scores.Practise this question →Read answer →