Machine Learning System Design Interview, issue 39, May 27, 2026
The Feature Space Trap
Senior ML Engineer interview at Netflix, and the interviewer asks:
“Your team engineered complex feature crosses that boosted offline AUC by 4%, but the platform team rejected the deployment because it violates our strict 20ms inference latency SLA. The team wants to scale the cluster. What do you do?”
Don’t say: “Just scale out the inference cluster, spin up larger instances, or use distributed caching to handle the increased load.”
Why chasing a 4% offline AUC boost can silently destroy your online inference SLA, and how to prune combinatorial bloat before funding cloud provider margins.
The full answer, with the mechanism and the arithmetic, is for paid subscribers on Substack.