Uber Interview Questions — Page 31

Uber·Data Scientist·Technical Phone Screen

May 2026

Uber data scientist interview with a heavy SQL and pandas focus, all centered around a single user events table. Five questions total, ranging from basic window functions to rolling averages. Nothing too wild but the breadth in one session was a bit much.

For each user, write a SQL query to return the first product they purchased along with the purchase timestamp.
Write a SQL query to count how many distinct users made at least two purchases on the same calendar day.
Find the top 3 products by total number of purchases using SQL.
Calculate a 7-day rolling average of total daily purchases across all users in SQL.
Using pandas, compute daily active users from the same dataset, defined as the count of unique user IDs per date.

“Pretty standard first-purchase problem.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber SWE interview that was basically one big deep-dive into building a parser and evaluator from scratch. More involved than I expected for a single session, they really wanted to see both a recursive and an iterative approach plus full error handling.

Design and implement a parser and evaluator for a parenthesized expression language that supports sum, prod, and set operators with lexical scoping. The input may be invalid, and your implementation must detect and report errors like mismatched parentheses, wrong arity, undefined variables, and integer overflow.
Walk through two implementation approaches for this evaluator: recursive descent versus an iterative stack-based approach. Compare their time and space complexity in terms of the input length.
How would you extend the set operator to support multiple variable bindings in a single expression, like (set x 1 y 2 body)?
How would you add a division operator with integer truncation toward zero and proper division-by-zero error handling?

“This was the whole interview basically.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

May 2026

System design round at Uber focused entirely on distributed consensus, specifically Paxos vs Raft. Pretty deep dive, felt more like a grad school oral exam than a typical eng interview.

Walk through how Paxos and Raft each achieve consensus over a replicated log. Compare their approaches to leader election, quorum requirements, log replication, membership changes, and behavior under network partitions.
Walk through specific failure scenarios: what happens when the leader crashes, when messages are delayed, or when a node loses its disk? How does the cluster recover in each case?
Given that operational simplicity and debuggability are priorities, which would you choose for a new service, Paxos or Raft, and why?

“This is where I spent most of the interview and also where I fumbled the most.”

View Post

Uber·Data Scientist·Onsite - Product Sense / Strategy

May 2026

A deep product and experimentation case for Uber's grocery vertical, framed around launching 'Eats Grocery' in NYC. The question was dense and multi-part, basically a mini product strategy exercise crammed into one prompt. Felt like a take-home disguised as a live interview.

For a delivery platform entering the grocery space in NYC, what makes grocery fundamentally different from restaurant delivery? Identify at least five distinct dimensions, then pick one growth bet and one efficiency bet, define success metrics and leading indicators for each, and propose a low-risk experiment to validate them. Also cover the top three risks and explain how experimentation constraints differ from the restaurant side.

“This was a lot.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Phone screen for a Software Engineer role at Uber, Airport & Travel team. The interviewer was actually from a different team out of New York, which felt a bit odd. The coding problem was a known one and I did not perform well on it.

Solve a graph/tree traversal problem that requires a DFS-based approach, with additional constraints that open up multiple optimization paths beyond the brute-force solution.

“I knew the brute-force DFS angle pretty quickly but then completely stalled when the interviewer pushed for optimizations.”

View Post

Uber·Data Scientist·Technical Phone Screen

May 2026

Uber DS interview that was basically one massive causal inference case study. The whole thing centered on rider incentives and marketplace effects, and it went deep fast. Walked out not totally sure how I did.

You're designing a rider-side discount offer targeted by a propensity model. What marketplace health metrics would you define to evaluate the program, and what are the formulas?
How would you estimate the causal effect of this incentive program under three different identification scenarios: randomized geo holdouts, a regression discontinuity design using offer score thresholds, and an instrumental variable approach using operational friction?
Write out the ROI formula for this incentive, accounting for cannibalization, subsidy burn, surge and ETA externalities, and long-term habit formation. How do you estimate LTV lift and amortize acquisition cost, and how do you put confidence intervals on the final number?
Design a two-stage randomization scheme at the city-week level and then the rider level to separately identify direct effects versus spillover effects. Define your estimands clearly.
With one primary metric, three key secondary metrics, and around twenty diagnostic checks, what multiple testing correction strategy would you use and why?
How would you pre-register and execute a heterogeneity analysis to detect effect moderation by city tier and weather conditions, while controlling for sign and magnitude errors?
What diagnostics would you run to validate the experiment: pre-trend checks, placebo tests, negative control outcomes? And what are your explicit criteria for declaring success versus triggering a rollback?

“I started rattling off conversion rate and GMV like a reflex, which was the wrong move.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Interviewed for a software engineering role at Uber. The main thing I remember was a follow-up about scaling to millions of records, which is pretty standard for Uber but still tripped me up a bit.

How would you scale your solution when the dataset grows to millions of records?

“This came as a follow-up after I'd already presented my initial solution, and I wasn't fully ready for it.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber SWE interview with a graph/topology problem that looked like a word puzzle but was really a cycle detection question in disguise. Not a bad experience, just needed to recognize what kind of problem it actually was faster.

You're given a list of words sorted in lexicographic order according to an unknown alien alphabet. Reconstruct a valid ordering of the letters, or return an empty string if no valid ordering exists.

“Took me a minute to see past the alien language flavor text and realize this is just topological sort.”

View Post

Uber·Machine Learning Engineer·Technical Phone Screen

May 2026

Uber ML Engineer interview for a pricing or marketplace team, centered almost entirely on a deep-dive project presentation. No surprises on format but the scope of what they expected you to cover in one project walkthrough was pretty intense.

Walk us through a recent ML project end-to-end: business problem, stakeholders, data, features, model pipeline, system architecture, evaluation, experimentation, launch decisions, monitoring, and business impact.
How is your project relevant to pricing, marketplace optimization, ranking, forecasting, causal inference, or business metric optimization?

“This is basically the whole interview.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

May 2026

Uber system design round for a software engineer role. The whole thing was one long deep-dive into a meeting scheduling system, and they really did go through every layer from data model to concurrency to capacity math.

Design a meeting scheduling system for a large organization, including APIs for creating, updating, and cancelling both single and recurring meetings.
How would you prevent double-booking of attendees and rooms, and how does your conflict detection algorithm work under concurrent requests?
How would you implement a free/busy query to find the earliest available time slot across N participants, accounting for time zones and room constraints?
Walk through your data model: what tables or collections would you use, what are the key fields, and would you choose SQL or NoSQL here?
How would you shard and index this system for multi-tenant scale?
Describe the full cancellation flow: how do you remove a meeting, propagate notifications, free resources, and keep an audit trail, and can you get cancel operations to O(log n)?
What are your nonfunctional requirements: expected QPS, latency targets, caching strategy, consistency trade-offs, and back-of-the-envelope capacity estimates?
How do you handle idempotency, deduplication, and retry logic during create, update, and cancel operations?

“The recurring meetings part is where I started to drift.”

View Post