Design an end-to-end recommendation system evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Design a movie recommendation system end to end. Clarify objectives (e.g., CTR, watch time, diversity), key signals (explicit ratings, implicit interactions), cold-start strategy, and a first modeling approach such as matrix factorization/NMF for user–item decomposition. Outline feature engineering, training data creation, offline evaluation (MAP/NDCG) and online A/B testing, freshness/real-time updates, and how you would roll out and monitor the system.
Quick Answer: Design an end-to-end recommendation system evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
hardSoftware EngineerTechnical ScreenML System Design
9
0
Design an end-to-end recommendation system
System Design Prompt: End-to-End Movie Recommendation System
You are tasked with designing an end-to-end movie recommendation system for a large-scale consumer platform. Assume a web/mobile product with millions of users and a catalog of tens of thousands of titles. Optimize for both user satisfaction and business impact under typical production constraints (latency, scale, and privacy).
Objectives
Clarify and prioritize measurable objectives, for example:
Engagement: CTR, play-start rate, watch time, completion rate
Satisfaction: rating after watch, thumbs up/down, long-term retention
List quality: diversity/novelty/serendipity, coverage
Business constraints: content promotion, age/region eligibility, licensing
Embeddings: text/video/image embeddings for content similarity
User attributes: new vs returning, inferred preferences
Cold-Start Strategy
Describe strategies for:
New users: lightweight onboarding, popular/trending, contextual bandits
New items: content-based similarity from metadata/embeddings, controlled exploration
Modeling (First Cut)
Propose a first modeling approach such as matrix factorization/NMF for user–item decomposition. State the objective, training procedure (e.g., ALS), and how it integrates into a two-stage system (candidate retrieval + ranking).
Feature Engineering
Outline features for retrieval and ranking:
User, item, and interaction features
Sequence/recency features
Cross features and constraints (eligibility, business rules)
Training Data Creation
Explain how to construct labeled datasets from logs, including: