Interview questions

Ensemble Learning Trade-offs and Why Bagging Reduces Overfitting

Companies: Pinterest

ML Fundamentals · Supervised Learning

Estimate b when features exceed samples

Companies: Google

ML Fundamentals · Regression & Statistical Modeling

Estimate heterogeneous treatment effects with causal ML

Companies: TikTok

Estimate OLS via streaming sufficient statistics

Companies: Citadel

Regression & Statistical Modeling

Estimate Volatility from Gaussian and Brownian Samples

Companies: Tudor

Evaluate and monitor a credit risk model

Companies: Capital One

Model Evaluation · Model Monitoring & Drift

Evaluate Classifier with Precision, Recall, and Fairness Metrics

Companies: Meta

Evaluate Ensemble Models for Bias-Variance, Speed, and Interpretability

Companies: Amazon

Evaluate Factors Before Replacing Recommendation Model

Companies: Meta

Evaluate Factual Accuracy and Consistency in a Domain-Specific RAG Chatbot

Companies: C3 Ai

LLM Evaluation · Retrieval-Augmented Generation

Evaluate Fake-Account Classifier with Precision and Recall Metrics

Companies: Meta

Evaluate Guangzhou performance with limited data

Companies: Weride

Evaluate K-Fold Cross-Validation for Model Selection

Companies: Walmart Labs

Evaluate Models for Credit-Risk Scoring at Capital One

Companies: Capital One

Evaluate New Model's Performance Against Existing System

Companies: Meta

Evaluate NLP Classification Models

Companies: Amazon

Model Evaluation

Evaluate Noisy Data for LLM Post-Training

Companies: Mercor

Annotation & Label Quality · Fine-Tuning & Alignment Techniques

Evaluate OutlierHandler Class for Code Quality and Testing

Companies: Capital One

Evaluate Product-Ranking Algorithm with Precision and Recall Metrics

Companies: Meta

Evaluate Promotions for Uber Eats Users

Companies: Uber

Experiment Design (A/B Testing)

Evaluate Python Class Design in Data Pipeline

Companies: Capital One