Uber Data Analyst Interview Guide

{"experience":"The hardest part for me was that the questions kept coming back to the same theme from different angles: how do you think about metrics, and how do you defend your analysis? My process started with a recruiter call where they asked the usual why this role and why Uber, but they also spent time walking me through the interview flow so I knew what to expect. After that, I had a panel round with two technical-style interviews. One was coding-focused and mixed Pandas with easy DSA, and the other centered on product metrics and A/B testing. The metric questions were very practical, like what could explain stable SLA but rising customer complaints, how I’d identify operational bottlenecks in a dataset, what metrics I’d track for a delivery or logistics system, how I’d segment users for analysis, and how I’d define a north-star metric. They also wanted me to talk through root cause analysis and how I’d validate insights before presenting them, and they asked everything one by one rather than in a rapid-fire style. The final stage was another panel, this time four interviews back to back. Two were behavioral and past-experience heavy, with a lot of tell me about a time questions. The other two were more technical: one DSA medium coding round and one more A/B testing discussion. Compared with other companies, the process felt shorter overall, but the final round had a take-home component that took longer than I expected. In that part, I had to walk through my findings and defend my logic, which made it feel less like a presentation and more like a live review of how I reasoned. Overall it was pretty standard, but the emphasis on metrics, experimentation, and explaining tradeoffs was very real. I ended up getting the offer, and I’d say the best prep is to practice explaining metric changes clearly, especially around complaints vs. SLA, and to be ready to defend a take-home analysis line by line. outcome":"Accepted offer outcome_color":"green prep_tip":"Practice explaining metric shifts like stable SLA but rising complaints, and be ready to defend a take-home analysis step by step. Also review Pandas plus easy-to-medium DSA and basic A/B testing since those came up directly in the panels."}