Interview questions

Decide Whether an LSTM Is Appropriate for Sequence Data

Companies: Citadel

Decision Trees in Mathematical Terms, and What Random Forests and XGBoost Improve

Companies: Squarepoint

ML Fundamentals · Supervised Learning

Deep-dive XGBoost handling and overfitting

Companies: Capital One

Model Evaluation · Supervised Learning

Defend a Quantitative Factor Research Project

Companies: Point72

Portfolio Construction & Alpha Research · Project Deep Dive

Defend Your Own Research Paper: Experimental Setup, Practicality and Later Work

Companies: Cohere

Model Evaluation · Project Deep Dive

Define scalable train/validation for churn

Companies: Hbo

Deploy multi-armed bandits safely

Companies: Meta

Derive and compare core ML and RL methods

Companies: Amazon

ML Fundamentals · Reinforcement Learning

Derive and regularize logistic regression

General practice

Derive correlation bounds and omitted-variable bias

Companies: Two Sigma

Probability · Regression & Statistical Modeling

Derive expected inversions and mean distribution

Companies: Hudson River Trading

Derive Linear Regression Solution

Companies: Meta

Derive logistic regression objective and gradients

Companies: Upstart

ML Fundamentals · Supervised Learning

Derive MLP shapes and explain PyTorch broadcasting

Companies: NVIDIA

Describe algorithm to find function maximum

Companies: Morgan Stanley

Describe Building and Deploying a Machine Learning Model

Companies: First American

Describe Your Machine Learning Project Experience

Companies: Voleon

ML Fundamentals · Project Deep Dive

Design a battery-life predictor and cold-start strategy

Companies: Google

Feature Engineering · ML System Design

Design a Churn Model: Handle Missing Data and Justify

Companies: Amazon

Design a Cold-Start-Aware Recommender

Companies: Yelp

Design a CVR model for RTB bidding

Companies: The Trade Desk

CTR Prediction & Ads ML · Feature Engineering