Interview questions

Improve classifier with noisy multi-annotator labels

Companies: OpenAI

Annotation & Label Quality · Model Evaluation

Last reported Feb 2026

Design an OOD detection system

Companies: OpenAI

Model Evaluation · Model Monitoring & Drift

Last reported Dec 2025

Train a classifier and analyze dataset

Companies: OpenAI

Model Evaluation · Supervised Learning

Last reported Sep 2025

Debug transformer and train classifier

Companies: OpenAI

ML Coding · Model Evaluation

Last reported Aug 2025

Debug a failing ML classifier

Companies: OpenAI

ML Fundamentals · Model Evaluation

Last reported Jul 2025

Design a response-ranking ML system

Companies: OpenAI

Model Evaluation · Ranking Models

Last reported Jul 2025

Answer ML fundamentals and diagnostics questions

Companies: TikTok

ML Fundamentals · Model Evaluation

APPLIED SCIENTIST

Companies: Amazon, eBay, OpenAI

ML Coding · Model Evaluation · Natural Language Processing

Build a Payment Fraud Detection Model

Companies: Adyen

ML System Design · Model Evaluation

Build and assess CTR prediction

Companies: Uber

CTR Prediction & Ads ML · Model Evaluation

Build and Defend a Baseline Model from a CSV

Companies: Microsoft

AI-Assisted Development · Model Evaluation

Build and evaluate airline delay prediction model

Companies: Capital One

Feature Engineering · Model Evaluation

Build and evaluate an order prediction model

Companies: Airbnb

Model Evaluation · Supervised Learning

Can One k Always Maximize kNN Leave-One-Out Accuracy?

Companies: C3 Ai

Model Evaluation · Supervised Learning

Choose Labels and Losses for Multiple Engagement Outcomes

Companies: Snapchat

Model Evaluation · Supervised Learning

Choose models for trading tasks

Companies: Citadel

ML Fundamentals · Model Evaluation

Choose Precision, Recall, or F1 for Agent Evaluation

Companies: ByteDance

Model Evaluation

Compare Logistic Regression and Random Forest in Python

Companies: Capital One

Model Evaluation · Supervised Learning

Compare Predictive Models from Two Vendors

Companies: Google

Model Evaluation · Sampling & Estimation

Compare two rare-event detection models statistically

Companies: Waymo

Hypothesis Testing · Model Evaluation

Compute and plot a precision–recall curve

Companies: Microsoft

ML Coding · Model Evaluation

Compute Recall and FPR from Confusion Matrices to Select Classifiers

Companies: Pinterest

Model Evaluation