All interviews
Netflix logo

Netflix

Senior

Netflix Ads Decisioning & Optimization engineer — real-time ad ranking, bidding, and pacing infra with sub-20ms ML serving

A senior (7+ years) distributed systems role on Netflix Ads' Decisioning & Optimization team, building the real-time ad decisioning path — ranking, scoring, bidding, and pacing — plus ML model serving infrastructure supporting dozens of concurrent hot-path models at sub-20ms P99 inference. Requires genuine ad-tech domain experience (2+ years) alongside distributed systems depth.

Step into this interview

Free · a live voice mock calibrated to this exact role

Practice this interview

What this interview tests

  • Real-time ad decisioning: ranking, scoring, bidding, pacing under strict latency/throughput SLAs
  • ML model serving infrastructure at sub-20ms P99 (routing, fallback, calibration, lifecycle)
  • Ad-tech domain fundamentals: inventory management, frequency capping, supply-demand
  • Auction mechanics and budget pacing/delivery optimization
  • Simulation/offline validation frameworks for marketplace changes
  • Operational excellence: reliability, observability, incident response

Common question themes

Design a real-time ad ranking/bidding system meeting strict P99 latency budgets

How would you build ML model serving infra supporting dozens of concurrent hot-path models with fallback tiers

Explain auction mechanics you've implemented or reasoned about: first-price vs second-price, reserve pricing, bid shading

How would you design a budget pacing system to keep campaign delivery accurate across a campaign's lifetime

Describe productionizing a data science model into a low-latency serving path

How would you build a simulation framework to validate marketplace changes before live rollout

View the original posting

Related interviews