Question bank

4,310 interview questions, answered.

Reference answers, what the interviewer is really testing, how it is graded, and the follow-ups that come next.

Easy 190Medium 2,498Hard 982
FiltersDifficulty, topic, company

4,310 questions

  1. 3193Clay hiring manager: GTM workflow design + platform fluencyMediumProduct
  2. 3194Constitutional AI, prompt injections, guardrailing (Anthropic SE)HardProduct
  3. 3195CSM interview — long-term value delivery and expansionHardProduct
  4. 3196Customer wants a feature that would compromise data governance — push backHardProduct
  5. 3197Customer's IT team wants to deploy in their VPC but won't give production credentialsMediumProduct
  6. 3198Define success metrics for a new AI productMediumProduct
  7. 3199Delivering bad news to a customerMediumProduct
  8. 3200Design a disaster-response coordination systemHardProduct
  9. 3201Design an AI agent for a streaming serviceMediumProduct
  10. 3202Design an AI-driven outbound motionHardProduct
  11. 3203Design an enrichment pipelineMediumProduct
  12. 3204Design an eval strategy for a new LLM featureHardProduct
  13. 3205Design for a model failure modeHardProduct
  14. 3206Diagnose high latency in an LLM inference pipelineMediumProduct
  15. 3207Disagreeing with a customer and holding the lineHardProduct
  16. 3208Distributed inference at high request volumeHardProduct
  17. 3209Emergency bed allocation for a 1100-bed hospital (Palantir TGH deployment)HardProduct
  18. 3210Explain why your RAG system can't guarantee 100% accuracy to a non-technical VPHardProduct
  19. 3211Fine-tune vs. RAG vs. prompt engineeringMediumProduct
  20. 3212GPU scheduling and batching for LLM inference workloadsHardProduct
  21. 3213Guardrails for production LLM appsMediumProduct
  22. 3214Healthcare client adoption is at 12% after 90 daysMediumProduct
  23. 3215How do you know your AI system is actually working well?HardProduct
  24. 3216Launch an AI feature into an existing productHardProduct