Design nearby place recommendations
Overview
This question evaluates skills in designing real-time, large-scale machine learning recommender systems that combine geospatial candidate generation, feature engineering, ranking models, personalization, privacy controls, fraud detection, and operational constraints like latency and throughput.
Meta
Sep 6, 2025, 12:00 AM
Machine Learning Engineer
Onsite
ML System Design
0
0
Real‑Time Nearby Places Recommendation System
Context
Design a mobile feature that recommends nearby places (e.g., restaurants, shops, attractions) to a user in real time as they move. The system should work globally, handle dense urban and sparse rural areas, respect user privacy, and scale to high QPS.
Requirements
- Define objectives and success metrics.
- Describe data sources: maps/POIs and user behavior.
- Candidate generation using geospatial filtering.
- Feature engineering: distance, popularity, personalization, context (time/weather).
- Ranking model choice and training signals.
- Cold‑start strategies for new users and new places.
- Exploration vs. exploitation strategy.
- Spam/fraud filtering.
- Privacy considerations.
- Latency/throughput budgets and system architecture.
- Online A/B testing and monitoring plan.
Solution
Login required
Show