System & Pipeline Design

Whether you can take a vague requirement, state assumptions and scale, and reason about trade-offs: batch vs streaming, storage choice, partitioning, idempotency, exactly-once, late/duplicate data, cost, and observability. Senior leveling is usually decided here.

• Use one repeatable frame: requirements → ingestion → storage → processing → serving → monitoring.
• Always cover late-arriving and duplicate data, idempotent loads, and SLAs.
• Know batch vs streaming trade-offs and when each is justified.
• Practice one domain pipeline end to end (clickstream, payments, or telemetry).

• Designing only the happy path (no failure, late data, or monitoring).
• Jumping to tools before stating requirements and scale.
• Flashy architecture with no fault tolerance, consistency, or cost story.