Uber Analytics & Experimentation interview questions focus on experimentation at scale inside a two‑sided marketplace where small measurement mistakes can have big business consequences. Interviewers typically evaluate your ability to design rigorous A/B tests and causal analyses (unit of randomization, sample size, guardrail metrics, and variance‑reduction), your statistical intuition for significance and power, and your product and operational judgment about interference, ramping and rollback. Expect a mix of case-style experiment design prompts, metric-definition and root‑cause scenarios, and hands‑on questions that probe your SQL/stats fluency and ability to interpret noisy results. For interview preparation, emphasize experiment design fundamentals, common pitfalls (SRM, interference, peeking, non‑normal metrics), and clear communication of assumptions and tradeoffs. Practice framing goals, choosing primary and guardrail metrics, sketching sample‑size calculations, and describing rollout plans and safety checks. Walk through a few real or mock investigations end‑to‑end—hypothesis to analysis to recommendation—so you can explain choices concisely to product and engineering partners under time pressure.