12 classification interview questions, asked at Amazon, Apple, Meta and 6 more companies. Last asked Sep 2026.
+1
Design a harmful content detection systemML designDetect harmful posts (hate, violence, self-harm, spam) across text and images, with high recall on severe harms and a path to human review.
Last asked Sep 2026
Predict employee attrition and explain what drives itML designYou get an HR dataset with one row per employee (age, job role, monthly income, overtime and similar columns) and a label saying whether the person left. Build a model that…
Last asked Aug 2026
Choose storage tiers from predicted file accessML designA cloud storage service can keep each user file in a fast, expensive tier or a cheap, slow archive tier, and moving data or reading from the archive costs money. Decide which tier…
Classify a stream of tokens with embeddingsML designGiven data that arrives as a stream, process the tokens online, turn them into embeddings, and build a model that classifies them. Cover how you process data as it arrives, which…
Last asked Jan 2026
Classification model lifecycle and click-through predictionML designWalk through every step of building a classification model and the techniques at each step: framing and labels, data collection and cleaning, features, model choice, training…
Last asked Dec 2025
Design a model that predicts posts a user will dislikeML designDesign a model for a news feed that predicts which posts a user would dislike or hide, so ranking can show fewer of them. Cover the labels (explicit negative feedback is rare)…
Last asked Sep 2025
Design an email spam detection systemML designDesign a system that flags spam email before it reaches the inbox. Cover the features (content, sender, links, behavior), the model, labels from user reports, adversaries who…
Last asked Aug 2025
Design a customer churn prediction systemML designPredict which business customers will stop using the product in the next 30 days, so account teams can act in time.
+1
Design a bot detection systemML designSeparate automated traffic from real users on a large website, in real time, while bots adapt to whatever you deploy.