Design Recommendation Infrastructure for microservices
Last updated: January 19, 2026
Quick Overview
Design a low-latency recommendation system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
DE Shaw
System Design
Software Engineer
DE Shaw
January 19, 2026Software Engineer
Onsite
System Design
Medium
48
13
3,249 solved
Design a low-latency recommendation system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
DE Shaw asks this during the Onsite to assess your understanding of the full ML lifecycle. They want to see how you translate a business problem into an ML objective, design the feature pipeline, and plan for model monitoring and retraining.
What the Interviewer Expects
- Define clear ML objectives with appropriate loss functions and metrics
- Design a comprehensive feature engineering pipeline
- Discuss model selection with trade-offs (complexity vs interpretability vs latency)
- Plan online and offline evaluation strategies including A/B testing
- Address serving infrastructure: batch vs real-time, latency requirements
- Consider data quality, labeling strategy, and feedback loops
Key Topics to Cover
ML objective formulation and metric selection
Feedback loops and model retraining
Model serving and latency optimization
Model selection and architecture
Online vs offline evaluation
Monitoring and model degradation detection
How to Approach This
- Start by clarifying functional and non-functional requirements with the interviewer.
- Estimate the scale: QPS, storage, bandwidth. This drives your design decisions.
- Draw a high-level architecture first, then deep dive into 1-2 critical components.
- Discuss trade-offs explicitly (e.g., consistency vs availability, SQL vs NoSQL).
- Address failure scenarios, monitoring, and how the system handles 10x traffic spikes.
Possible Follow-up Questions
- How would you debug a model that works well offline but poorly online?
- What would you do if model performance degrades over time?
- How would you run A/B tests on different model versions?
- How would you handle a 10x increase in prediction requests?
Practice a Similar Problem on Codemia
Solve a related problem with our interactive workspace, get AI feedback, and view detailed solutions.
Solve on CodemiaSample Answer
Requirements
Functional Requirements
- Real-time Recommendation Generation: The system must provide personalized recommendations to users with low latency (under 100ms).
- User Feedback Loop: Capture...
Capacity Estimation
To estimate capacity, assume the following:
- User base: 10 million active monthly users.
- Requests per user: Average of 5 recommendations per session, with 2 sessions per day.
- **Total requ...
Submit Your Answer
Markdown supported