Machine Learning System Design Interview, issue 31, May 19, 2026

The Real-Time Pricing Paradox

Senior ML Engineer interview at Amazon Go, and the interviewer asks:

Your team just built a Kafka-backed dynamic pricing model for our physical grocery stores with sub-10-millisecond feature freshness. But the Director of Retail Operations immediately rips it out and mandates day-old batch processing. Why?

Don’t say: It must be an infrastructure bottleneck. The Flink cluster is probably OOMing, or the downstream feature store can’t handle the high QPS writes, so they forced a fallback to batch to save on compute costs.

How over-engineering streaming pipelines silently nukes edge network bandwidth and breaks customer psychology, and why a 2:00 AM batch materialized view is the true elite-level solution.

The full answer, with the mechanism and the arithmetic, is for paid subscribers on Substack.

Read it on Substack

Get the next one

Free on Substack. Unsubscribe in one click.