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 0Medium 2Hard 3
Filters · onDifficulty, topic, company

Difficulty

Clear all

24 questions

  1. 1Bagging vs boosting vs Random ForestData· Amazon
  2. 2BiLSTM vs TransformerMediumData· Amazon
  3. 3Dimensionality reduction methodsData· Amazon
  4. 4Handling imbalanced dataData· Amazon
  5. 5Bias-variance trade-offData· Amazon
  6. 6Company vs client interests misalignedData· Amazon
  7. 7Delivered something customers didn't know they neededHardData· Amazon
  8. 8Design and evaluate a book recommenderData· Amazon
  9. 9Design and evaluate an A/B testData· Amazon
  10. 10Missing a long-term commitment (Ownership)Data· Amazon
  11. 11Outside-the-box simplification (Invent & Simplify)Data· Amazon
  12. 12Prove logistic loss has a global minimumData· Amazon
  13. 13Technical deep-dive on an issue (Dive Deep)HardData· Amazon
  14. 14The project you're most proud ofData· Amazon
  15. 15Urgent decision with incomplete information (Bias for Action)MediumData· Amazon
  16. 16Classify products into categoriesData· Amazon
  17. 17Closed-form linear regression vs gradient descentData· Amazon
  18. 18How did you debug your ML project?HardData· Amazon
  19. 19KL divergence vs cross-entropyData· Amazon
  20. 20Statistical methods you've actually usedData· Amazon
  21. 21Supervised vs unsupervised vs reinforcement learningData· Amazon
  22. 22The 15-20 minute science talkData· Amazon
  23. 23Why that loss function?Data· Amazon
  24. 24Your favorite ML paperData· Amazon