Game / Social Recommendation System Design

Problem Overview

Design a recommendation system for games (or content) on Roblox. Variants observed across candidates: 1. Full ML pipeline: How do you recommend games to users? Cover the entire ML/recommendation pipeline — feature selection and engineering, model architecture, training data, loss functions, optimization algorithms, offline/online evaluation metrics, and A/B testing. Not…

• 7 candidate-reported follow-ups — the exact probes interviewers asked, with the trigger for each
• The rest of the problem statement — full requirements, constraints, and edge cases
• Approach and trade-offs — what passing candidates did, and the mistakes that sink people