Your question is Scalable Data Ingestion for AI. 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 an AI application that processes both structured transactional data and unstructured images?
Explain a practical hybrid batch and streaming architecture using specific technologies. Cover source ingestion, image and metadata storage, schema evolution, validation, deduplication, lineage, orchestration, replay and backfilling, security, and serving data to downstream AI workflows. State your assumptions and describe how you would monitor freshness, completeness, processing failures, and cost without coupling image binaries to transactional tables.
