How the interviewer opens it
“Design a system for traffic from a one-time event, such as a Super Bowl advertisement. The service must remain reliable during the event.”
The prompt is vague on purpose. The first minutes are for you. Reduce the prompt to the core features and agree on the numbers before you draw.
What to clarify first
- Visitor flow
- Audience and timing
- Read, write, and retry traffic
- Payload size and retention
- Latency and success targets
- Reliability and consistency
- Submission outcomes
- Receipt meaning
- Failure boundary
- Exclusions and abuse
The interviewer has the answers. In a SysRound round, Alex gives these answers, as a real interviewer does.
Scope of a strong answer
Core requirements
- Show a promotion page
- Accept contact details and return an entry receipt
- Return the status of a previous submission
Usually out of scope
- Payments and account access
- Limited prize inventory and prize allocation
- Advanced fraud detection
- Continued service after a full region fails
What interviewers ask about
- The first minute
- Millions of visitors arrive faster than new servers can start. Page access, entry acceptance, status checks, and retries compete for different resources.
- Submission outcomes
- A timeout does not show whether the service stored an entry. A false success or a duplicate breaks the receipt contract.
- Before the advertisement
- The event offers one chance to serve the audience. A capacity estimate does not prove that every dependency can handle the peak.
Common mistakes
- Assumes the product needs payments or scarce inventory without questions.
- Optimizes response time while allowing acknowledged entries to disappear.
- Uses average traffic instead of peak traffic and omits status checks or retries.
- Relies on autoscaling without a capacity test or quota check.
- Adds a queue without a clear acceptance point.
- Builds a complex permanent platform for a short event without a cost or reliability reason.
The bar by level
- Senior (L5/E5)
- Calculates the complete peak workload and survives one zone failure. Defines acceptance, timeout behavior, and overload limits. Proposes a realistic test before the event.
- Staff (L6+/E6+)
- Connects the success target to failure cases and capacity evidence. Tests retry assumptions and dependency limits. Defines event readiness, recovery plans, and the regional failure boundary.