Typical flow: recruiter screen → technical screen (SQL + experimentation/recsys) → onsite loop with case studies, product/analytics deep dives, ML/statistics, and behavioral culture fit. Some roles include a short modeling or SQL exercise.

SQL, Python or R, statistical inference, experimental design and analysis, recommendation/ranking methods, metrics like NDCG/MAP, and the ability to turn insights into product decisions. Experience with large‑scale data pipelines is a plus.

Common themes: improving homepage personalization, diagnosing a drop in engagement, designing robust A/B tests with guardrails, evaluating content valuation/forecasting, and prioritizing features with measurable member impact.

Not strictly required, but familiarity with streaming metrics (starts, completes, time watched), churn/retention drivers, and content lifecycle analytics helps you reason about trade‑offs quickly.

Netflix aims for top‑of‑market pay. Mid‑senior roles commonly include a high base salary with smaller equity components compared to peers. Exact numbers vary by level, location, and experience.

Expect probing on metric selection (primary vs guardrails), variance reduction (CUPED / stratification), power trade-offs, sequential testing risks, heterogeneous treatment effects, and interpreting conflicting metric movements. You should articulate when to stop an experiment early and how to validate external validity.

Diversity/novelty (intra-list diversity), discovery uplift, retention contribution, title completion ratio, early session abandonment, satisfaction proxies (downstreams like churn reduction). Show awareness of metric gaming and long-term engagement vs short-term click uplift.

High for personalization & product decisions: difference-in-differences for staggered rollouts, synthetic controls for content launches, instrumental variables for selection bias, uplift modeling for targeted messaging. Be ready to explain assumptions and failure modes.

Examples: diagnose drop in retention after UI change, design experiment for recommendations ranking, compute and interpret NDCG change, build causal approach for content valuation, propose cold-start strategy metrics, analyze multi-armed bandit trade-offs, segment churn drivers for new markets.