AppliedAIPrep logoAppliedAI/Prep
← SQL & Data Engineering / 54
hardDatabricksSnowflakeGoogle

How do you handle late-arriving data in a streaming or incremental pipeline?

Events show up minutes or days after they happened, and a window that already closed gives wrong counts. The signal is event-time vs processing-time, watermarks with allowed lateness, and how you correct already-emitted aggregates.

Updated Aug 2026 · Grounded in real Applied AI Engineer interview loops and written to a senior-engineer editorial bar.

Events show up minutes or days after they happened, and a window that already closed gives wrong counts. The signal is event-time vs processing-time, watermarks with allowed lateness, and how you correct already-emitted aggregates.

Unlock the other 750 answers · ₹2,000 / $25Your progress and mastery stay saved · 6 months · one payment · no auto-renew
UP NEXT ON YOUR JOURNEY
Next in this trackYou shipped a logic bug three months ago. How do you safely backfill and reprocess the affected data?Next in this trackHow do you query deeply nested JSON with arrays in SQL, and when do you flatten vs keep it nested?Next in this trackCompare CDC variants: log-based, query-based, and trigger-based. What are the failure modes of each?
DISCUSSION · 0

No comments yet — be the first to share your approach.