← Back to Directory

Uber interview notes, page 36 of 61

Large Enterprises

Most of these Uber write-ups come from Software Engineers (407), then Data Scientists (78) and Machine Learning Engineers (65). The most reported rounds are the technical phone screen (387), the system design round (95), the product sense round (46) and the onsite coding round (23).

609 interview notes · updated Aug 2026

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber software engineer interview with a classic grid traversal problem. Nothing too exotic but it's the kind of question where you either know BFS/DFS cold or you don't, and the pressure of an actual interview has a way of making you second-guess yourself.
  • Given a 2D binary grid representing a map of land ('1') and water ('0'), count the number of islands. An island is a group of horizontally or vertically connected land cells, and the grid is surrounded by water on all edges.

“I went with DFS and it worked fine, but I wasted probably two minutes overthinking whether BFS would be more impressive.”

View Post

Uber·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Uber PM interview with a market sizing question on food delivery. Pretty standard case interview feel, nothing too wild.
  • Estimate the size of the food delivery market in the United States.

“I started top-down and then tried to sanity check with a bottom-up pass, which I think helped.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Did a coding round for a SWE role at Uber. One question, classic-ish number manipulation problem. Not a lot of context to share but it was the kind of thing that feels easy until you're actually writing it out.
  • Given a number, find the closest palindrome to it.

“I thought I had this figured out in the first two minutes and then spent the next fifteen realizing I did not.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber SWE interview focused on a Union Find problem involving ride-sharing logs. The main coding question wasn't too bad but the follow-ups kept coming and that's where things got interesting.
  • Given plain-text ride-sharing logs (each entry has a timestamp and indicates that user A shared a ride with user B) and a list of all users, find the earliest timestamp at which all users become connected.
  • How would you extend the solution to handle multiple types of log entries, not just ride-sharing events?
  • What is the approximate time complexity of your solution?

“Went in having seen something similar on a prep forum so I had a rough idea.”

View Post

Uber·Data Scientist·Technical Phone Screen

Jun 2026
Deep dive into marketplace experimentation for Uber's pricing team. The whole interview was basically one long case study about A/B testing a new pricing model for a ride-hailing platform, with follow-ups stacked on top of each other. Solid question set but it gets pretty granular fast.
  • Before designing the pricing experiment, how would you think through a driver's mental state when they receive a trip request prompt?
  • What primary metrics and guardrail metrics would you define for riders, drivers, and the overall marketplace when testing a new pricing model?
  • If you run the experiment in a single city with standard treatment and control groups, what interference problems could arise from marketplace network effects?
  • How would you design a switchback experiment for a single city to avoid those interference issues?
  • How would you handle limited sample size in a switchback test?
  • What are the tradeoffs of using full-day switchback windows versus shorter windows?
  • What are the tradeoffs between using a larger versus smaller treatment group in this kind of experiment?
  • What are the tradeoffs between longer and shorter treatment windows in a switchback design?
  • If the experiment is later expanded to 12 cities, how would you redesign the experiment to increase sample size and improve validity?

“This was a weird opener and I didn't expect it.”

View Post

Uber·Data Scientist·Hiring Manager Screen

Jun 2026
Behavioral screen for a Data Scientist role at Uber, focused on leadership potential and cultural fit. The hiring manager wanted to see how you think about your own work and whether you can operate without a title backing you up.
  • Walk me through your resume and highlight the project you are most proud of.
  • Tell me about a time you received critical feedback. How did you respond?
  • Describe a situation where you had to influence stakeholders without having formal authority over them.

“I picked a project with decent numbers and talked through it as situation-action-result.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber SWE interview with a graph problem that looks manageable until you realize the rerooting trick is what they're actually testing for.
  • Given a directed graph that forms a tree when treated as undirected, find the minimum number of edges that need to be reversed so every node can reach node 0.

“My first instinct was just BFS from 0 and count wrong-direction edges.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber SWE coding round, pretty standard array problem but the follow-up tripped me up more than I expected.
  • Given a sorted array of integers that may include negative numbers, return the squares of all elements in sorted order. Follow-up: find the k-th smallest squared value.

“The base problem I got through fine.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber SWE interview with a tree path problem that's basically a twist on a known LeetCode problem. Not the hardest round I've had but definitely requires you to know your bitmask tricks cold.
  • You're given a tree (as a list of edges) rooted at node 0, where each node holds a lowercase letter. For each query node, count how many paths from that node up to the root have a character multiset that can be rearranged into a palindrome (at most one character with odd frequency).

“I recognized the palindrome-via-bitmask pattern pretty quickly since I'd seen something similar before, but the node-vs-edge distinction tripped me up for a bit.”

View Post

Uber·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Uber MLE interview with a coding round focused on data structure design. The problem was more involved than a typical LeetCode question and pushed into system-ish territory, which I wasn't fully expecting.
  • Design and implement an in-memory revenue tracking system for a referral network, where each customer's aggregate revenue includes their own direct revenue plus all revenue from customers they referred (recursively). Support an insert operation that adds revenue and optionally links a referrer, and a query operation that returns the top K customers with aggregate revenue above a given threshold, sorted by revenue descending with ties broken by customer ID.

“Took me a few minutes to really internalize the recursive part.”

View Post