Design a high-throughput A/B Testing System
Last updated: May 4, 2026
Quick Overview
Design a high-throughput a/b testing system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
Doordash
System Design
Software Engineer
Doordash
May 4, 2026Software Engineer
Onsite
System Design
Easy
37
11
4,028 solved
Design a high-throughput a/b testing system that handles millions of requests. Discuss trade-offs in consistency, availability, and performance.
This ML system design question from Doordash's Onsite tests your ability to think about ML systems at scale. The interviewer expects discussion of data quality, feature stores, model serving infrastructure, and A/B testing strategy.
What the Interviewer Expects
- Map the business problem to a concrete ML objective
- Propose reasonable features and a baseline model
- Discuss basic model evaluation metrics
- Outline a simple serving architecture
Key Topics to Cover
Feature engineering and feature stores
Monitoring and model degradation detection
Feedback loops and model retraining
Online vs offline evaluation
A/B testing and experimentation
Training pipeline and infrastructure
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
- What would you do if model performance degrades over time?
- How would you debug a model that works well offline but poorly online?
- How would you handle a 10x increase in prediction requests?
- What is your model retraining strategy?
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:
- Experiment Creation: Users can create A/B tests with defined metrics, variants, and durations.
- Traffic Allocation: System should route traffic to different v...
Capacity Estimation
Back-of-Envelope Calculations:
- User Base: Assume DoorDash has 30 million active users.
- A/B Test Frequency: If we run 10 simultaneous A/B tests, each test targeting 1% of users, that’s ...
Submit Your Answer
Markdown supported