Your question is Scalable Data Ingestion Pipeline. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How would you design a scalable data ingestion pipeline for a client with legacy infrastructure at QuantumBlack?
Explain how you would assess existing systems, select batch or streaming ingestion patterns, and introduce modern components without disrupting current operations. Address schema management, incremental loading, idempotency, data quality, orchestration, observability, security, migration, and backfilling. State which requirements you would clarify before choosing throughput, latency, retention, and recovery targets.