Katkat

Topic

Machine Learning interview questions

572 Machine Learning questions from 133 companies, with hints, runnable code and solution walkthroughs.

Questions572

59 easy323 medium190 hard

Your progress0/572

Machine Learning interview questions by the numbers

Last updated

Katkat has 572 Machine Learning interview questions from 133 companies; the most come from Amazon (10%), OpenAI (6%) and Meta (5%). By difficulty: 10% easy, 56% medium and 33% hard. The most common round is Technical Screen (61%). Frequent concepts include Machine Learning, Feature Engineering, Model Evaluation and Optimization.

Companies (133)

Difficulty

  • Easy10% · 59
  • Medium56% · 323
  • Hard33% · 190

Interview rounds

  • Technical Screen61% · 348
  • Onsite28% · 163
  • Online Assessment8% · 46
  • HR Screen3% · 15

Roles

  • Data Scientist46% · 264
  • Machine Learning Engineer35% · 202
  • Software Engineer9% · 53
  • Applied Scientist5% · 26
  • Quantitative Researcher2% · 13

Frequent concepts

  • Machine Learning7% · 39
  • Feature Engineering5% · 29
  • Model Evaluation4% · 23
  • Optimization4% · 23
  • Deep Learning4% · 21
  • System Design3% · 18

Most mentioned in interview experiences

  1. 1.Answer core ML fundamentals questions2 reports
  2. 2.Compare deep learning framework trends2 reports
  3. 3.Evaluate Factual Accuracy and Consistency in a Domain-Specific RAG Chatbot2 reports
  4. 4.Explain Core ML Interview Concepts2 reports
  5. 5.Explain Overfitting and Transformer Attention2 reports
  6. 6.Explain SHAP in an ML System2 reports
  7. 7.Implement bagging with decision trees2 reports
  8. 8.Present and defend your data challenge end-to-end2 reports
  9. 9.Analyze Temperatures and Update Regression1 report
  10. 10.Analyze Town Temperatures and Select Predictors for NYC1 report

Counted from 122 candidate interview experiences on Katkat. How we collect and count

Showing 571–572 of 572 results

  • Explain Data Collection Choices and Their Impact on Downstream Model PerformanceMedium

    A machine learning screening question: explain how the data for a project on your resume was collected and processed, then discuss how data choices affect performance on downstream tasks. It tests first-hand knowledge of data pipelines, causal reasoning about data quality, and how to measure data effects.

    xAIMachine Learning EngineerMachine LearningOnsite
  • How Would You Prevent Hallucinations in an LLM-Based System?Medium

    This question evaluates a candidate's understanding of hallucination in large language models, including underlying causes, detection, and system-level mitigation approaches, as well as their ability to reason about model reliability and failure modes.

    xAISoftware EngineerMachine LearningHR Screen

More topics