Design a Distributed Ad Serving and Targeting Platform
Last updated: March 3, 2025
Quick Overview
Design real-time bidding infrastructure with user targeting, ad auction logic, impression tracking, and fraud detection under 10ms latency.
ByteDance
March 3, 20257
7
3,811 solved
Design real-time bidding infrastructure with user targeting, ad auction logic, impression tracking, and fraud detection under 10ms latency.
Advertising is ByteDance's primary revenue source. This tests ad tech system design with strict latency constraints.
What the Interviewer Expects
- Design real-time bidding with sub-10ms latency
- Implement user targeting with interest graphs
- Build ad auction logic with fairness constraints
- Design impression and click tracking pipeline
- Address ad fraud detection
Key Topics to Cover
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 do you balance ad relevance with advertiser bid amounts?
- How do you handle frequency capping across devices?
- What is your approach to measuring ad attribution?
Sharpen Your Skills on Codemia
Practice similar problems with our interactive workspace, get AI feedback, and track your progress.
Practice System Design ProblemsSample Answer
Serving Architecture
When a user opens the app, the ad server receives a request with user context (demographics, interests, session signals). The server queries the ad in...
Auction Design
Use a second-price auction: the winner pays one cent above the second-highest bid. This incentivizes truthful bidding. Apply quality score multipliers...
Submit Your Answer
