Uber Interview Questions — Page 45

Uber·Software Engineer·Technical Phone Screen

Jun 2026

Uber SWE interview, one meaty design-and-implement question about a hierarchical revenue structure. Pretty heavy on data structure justification, not just code.

Design a data structure that stores revenue entries in a parent-child hierarchy. It needs two insert operations: one for top-level entries and one that links an entry to a specific parent by revenue value. At any point it should support returning the top K smallest revenues across the whole collection. Walk through your choice of underlying structures, how you handle ties, and the time complexity of each operation.

“I went with a hash map for the parent relationships and a min-heap for the top-K queries, which felt right.”

View Post

Uber·Data Scientist·Technical Phone Screen

Jun 2026

SQL-heavy technical screen for a Data Scientist role at Uber. Two questions, both around the same schema, but the second one had enough edge cases to keep me busy for a while.

Given a users table with user_type and is_active columns, write a SQL query to return the total count of active riders and active drivers on the platform.
Using the same schema (trips, users, city tables), find the driver with the highest total earnings year-to-date for 2025. YTD is defined in each city's local timezone, trip timestamps are stored in UTC, and you need to handle ties by picking the smaller driver_id. Drivers with zero qualifying trips should still appear but shouldn't win unless everyone else is also zero. Provide two versions: one with window functions and one with GROUP BY plus ORDER BY LIMIT 1.

“Straightforward group-by on user_type where is_active = 1.”

View Post

Uber·Data Scientist·Technical Phone Screen

Jun 2026

Uber DS interview focused entirely on a single extended case around ETA estimation and marketplace experimentation. Seven connected sub-questions, all building on each other, which was actually a nice format but also meant one shaky answer snowballed into the next.

Which metrics would you track to evaluate the impact of lowering ETA? Propose a primary metric and relevant guardrail metrics, covering the rider funnel, cancellations, completed trips, gross bookings, marketplace health, and longer-term retention.
From a business perspective, why does reducing ETA matter to Uber?
Historical data shows a positive correlation between ETA and session conversion rate, meaning higher ETA correlates with more conversions. How do you interpret this, and what confounders or aggregation issues might explain it?
How would you design a causal experiment to measure the impact of lowering ETA? Walk through when a user-level A/B test works, when a switchback experiment over geography and time is better, and when you'd reach for synthetic control.
After the experiment, the 95% confidence interval on the lift in session conversion rate is [-5%, +1%]. How do you interpret this for decision-making?
How would you improve the precision and statistical power of the experiment?
The PM asks for a deep dive on users who took at least 5 trips during the experiment. How do you respond, and how do you communicate the statistical risk of this request?

“Started with session conversion rate as the primary, which felt right, then listed cancellation rate and driver utilization as guardrails.”

View Post

Uber·Software Engineer·Onsite - Behavioral / Leadership

Jun 2026

Behavioral round at Uber for a software engineering role, basically one big question about your best project dressed up with six sub-parts. Not a bad experience but it felt more like a structured debrief than a real conversation.

Walk me through the most impressive project you've worked on, covering the business problem, your role, the technical decisions you made, how you measured success, the hardest challenge you faced, and what you'd change if you did it over.

“This is a lot to hold in your head at once.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026

Uber SWE interview with a graph theory problem that started straightforward and then got extended mid-session. The follow-up felt like the real test.

Given a directed graph of n services with dependency edges, implement a function to detect whether the graph contains at least one cycle. Return true if a cycle exists, false otherwise.
Follow-up: instead of a single dependency list, you now receive multiple separate dependency arrays for the same set of services. Treat the union of all edges as one graph and determine whether any cycle exists. Describe your approach, data structures, and complexity.

“Classic DFS with a recursion stack to track back edges.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

Jun 2026

Uber system design round for a software engineer role. Just one question but it took the full hour and I left feeling like I'd only scratched the surface.

Design a messaging system like WhatsApp.

“I started with the usual stuff, users, messages, storage, but the scope kept expanding.”

View Post

Uber·Data Scientist·Onsite - System Design / Architecture

Jun 2026

Uber DS interview, system design focused. One big open-ended question about building an ETA system from scratch, covering basically everything: labels, features, models, metrics, monitoring, experiments. Felt more like a 45-minute architecture conversation than a typical ML interview.

Design an end-to-end ETA prediction system for a ride-hailing app covering both pickup and trip ETAs. Walk through prediction targets and label construction, feature selection, model choices, offline and online evaluation, and how you'd handle issues like metric tradeoffs, calibration, selection bias from canceled trips, traffic shocks, GPS noise, geospatial cold start, latency constraints, retraining, monitoring, and post-launch experimentation.

“This is a monster of a question and I underestimated how much ground they actually wanted covered.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026

Uber technical screen with a classic stack-based problem. Nothing too wild but it took me longer than I'd like to admit to get the logic right.

Given a list of log messages representing function start and end events on a single-threaded CPU, compute the exclusive execution time for each function by its ID.

“The tricky part isn't the stack mechanics, it's remembering to handle the timestamp math correctly when a function resumes after a nested call returns.”

View Post

Uber·Software Engineer·Onsite - Behavioral / Leadership

Jun 2026

Behavioral round at Uber for a software engineering role, basically one big question about your proudest project but with a lot of layers to it. Not a bad experience, just more structured than I expected.

Walk me through the project you're most proud of or consider your most impressive work. Cover the business and technical context, your specific contributions, the key technical decisions and trade-offs you made, the measurable impact, and what you'd do differently looking back.

“This sounds like one question but it's really five.”

View Post

Uber·Software Engineer·Online Assessment (OA)

Jun 2026

Uber SWE online assessment, one algorithmic problem centered on prefix sums and binary search. Pretty clean problem once you see the pattern, but the off-by-one stuff in the binary search bounds is where people apparently blow it.

Given a non-decreasing array of stall prices and a list of queries, each with a starting position and a budget, find the maximum number of items you can buy greedily moving rightward from that position without exceeding the budget.

“The greedy insight is pretty quick to get to since prices only go up, so you always want to buy left to right.”

View Post