Ziqin demonstrated strong clarity and initiative in breaking down the system design for a payment platform. The session covered key aspects including requirements, API design, high-level components, and scalability strategies.

Areas of strength included:

• Requirements clarity: Quickly enumerated the core functional and non-functional requirements, setting a strong foundation for the design.
• Domain modeling: Clearly identified key entities and correctly distinguished Payment from Transaction, demonstrating solid domain understanding.
• Early settlement awareness: Proactively included payment settlement as a first-class functional requirement rather than an afterthought.
• API and schema depth: Delivered a comprehensive API definition and began expanding entity schemas with appropriate fields and relationships.
• Scalability discussion: Addressed scale through sharding strategies and CDC-based state synchronization for downstream consumers.
  
  

Areas for improvement:

• Clarify whether the hold path is synchronous or asynchronous, and why. Synchronous solution is designed due to requirement of payment system.
• Explicitly call out latency constraints (e.g., ~2s due to external payment provider dependencies).
• Describe batch-based charging and Explicitly explain how batching is orchestrated and monitored:
  
  • Batch processing mechanics: How batches are created, tracked, retried, and reconciled.
  • Large-scale batches (e.g., 100k payments)
  • Batch file generation
  • Storage in object storage (e.g., S3)
  • Scheduled downstream processing (e.g., nightly at 10pm)

• Clarify durability guarantees (e.g., traceability of failed transactions, state recovery post-crash)
• Embrace more collaborative moments with your interviewer—for instance, naturally inferring steps like DB validation during payment creation
• Demonstrated some deep dive around data correctness guarantees — aim to confidently frame how at-least-once delivery + deduplication = exactly-once semantics for hold, charge and settlement phases
• Considered idempotency with a composite deduplication key strategy

Overall, Ziqin shows great structure and promise in system design interviews. With additional work on durability, correctness, and deeper tradeoff thinking , he will be well-positioned for success in senior-level interviews.

‍