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 3Medium 27Hard 13
Filters · onDifficulty, topic, company
43 questions
- 1Longest Increasing SubsequenceMediumCodingAI ML·
Atlassian, Meta - 2What's the effect of adjusting an LLM's context window size?MediumAI MLTechnical·
OpenAI - 3How would you build an object detection system to identify and locate flowers in images?HardAI ML·
Apple - 4You have a dataset comprising 1,000 avatar images and 100,000 user descriptions with associated avatar images. Create a model that recommends an image from a new set of 100,000 images for a user description.HardAI ML·
SAP - 5Define hallucinations in LLMs.MediumAI MLTechnical·
Anthropic, Mastercard - 6Design an evaluation framework for ads ranking.MediumAI MLProduct·
Meta - 7Design a fake news detection system.HardAI MLProduct·
Meta - 8Design an ML experiment tracking and analysis platform.HardAI MLSystem design·
Google - 9Design a personalized news ranking system.HardAI MLSystem design·
Meta - 10Design a product recommendation system.HardAI MLSystem design·
Meta, Pinterest - 11What metrics would you track to evaluate the performance of your ML pipeline?MediumAI MLSystem design· PAPerplexity AI
- 12How is gradient descent and model optimization used in linear regression?MediumTechnicalAI ML·
OpenAI - 13How would you handle hallucinations in a generative AI model deployed to users?MediumAI MLTechnical·
Anthropic, OpenAI - 14Implement a k-nearest neighbors algorithm.EasyCodingAI ML·
Amazon, Pinterest - 15Design an inference batching system for a single GPU that can handle up to 100 inputs per batch while users wait synchronously, maximizing utilization under compute constraints.HardAI MLSystem design·
Anthropic - 16You received user feedback that Yelp's restaurant recommendation is not personal enough. What would you do?MediumAI MLProduct·
Google - 17How do you stay up to date with advancements in machine learning?MediumBehavioralAI ML·
Anthropic, Capital One - 18What are the different types of machine learning?MediumTechnicalAI ML·
Google - 19What is overfitting or underfitting? Which models are most likely to experience this, and why?MediumTechnicalAI ML·
Nvidia, OpenAI - 20What are common linear regression problems?MediumDataTechnical·
Amazon - 21Why is a deep neural network better than a shallow neural network?MediumTechnicalAI ML·
Google, Nvidia - 22Design an app where customer listening history is collected after they have listened to a song for more than 30 seconds for analytics use.MediumAI MLSystem design·
Amazon - 23Design a system that recommends the next Confluence pages a user should view.HardAI MLSystem design·
Atlassian, PayPal - 24Design a system that filters waste by detecting paper and putting it in the correct bin.MediumAI MLSystem design·
Samsung