Real-time fraud scoring
Design a real-time fraud scoring system that must return an approve/decline/review decision within 100ms p99 inline in the payment authorization path, at ~10,000 transactions/sec. Decisions use velocity features (cards-per-device-in-1h, amount-per-account-in-24h), a model score, and rules. Features must be fresh (a card used 3 times in the last minute should reflect that). Describe the architecture, the feature store, and how you keep features both fast and fresh without blocking the auth path.
What a strong answer looks like
Clarify scale and constraints first. Propose a clean component breakdown, then go deep on the hard parts (data model, bottlenecks, consistency, failure modes) and name the trade-offs you are making.
Clarify5:30 left
Estimate5:30 planned
Design16:30 planned
Deep dive13:30 planned
Failure9:00 planned
Which questions mattered is sealed until you submit. Telling you now would just be handing over the edge cases.
Run or narrate your approach, then ask the coach.