← Back to Directory

Uber interview notes, page 37 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·Onsite - System Design / Architecture

Jun 2026
System design round at Uber for a backend role, centered entirely on designing a cart and checkout system for an on-demand delivery app. The prompt was intentionally left vague so you had to drive the requirements yourself, which I found more stressful than the actual design work.
  • Design the backend cart system for an on-demand delivery app, covering the full lifecycle from cart creation through checkout and fulfillment.
  • How would you model the data for a cart system, and what indexing strategy would you use to support common queries like fetching a user's active cart or all items in a cart?
  • Multiple devices can update the same cart simultaneously. How do you handle concurrent modifications and prevent conflicts or data corruption?
  • How do you ensure that checkout converts a cart to an order exactly once and that the cart state is valid at the time of conversion?

“The prompt is intentionally vague and they want you to ask clarifying questions before touching the design.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Phone screen for a Software Engineer role at Uber. The coding part was straightforward but the real interview was the design conversation that followed, which felt like it mattered a lot more than getting the code right.
  • Implement an LRU Cache with get and put operations in O(1) time.

“Finished the implementation in about 25 minutes, no major issues.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Bar Raiser round at Uber for a senior software engineer role, focused on architectural depth, technical disagreement, and ownership. Three heavy parts with follow-ups that probe for the uncomfortable stuff you'd rather not revisit.
  • Walk me through a complex system architecture you designed or led. What were the key trade-offs, what alternatives did you reject, and how did the system actually perform?
  • Tell me about a significant technical disagreement you had and how you resolved it. How did you drive adoption of a standard or practice across people who weren't on board initially?
  • Give me a specific example of Customer Obsession or Ownership where you went beyond your assigned scope to protect the user experience or take responsibility for something no one asked you to own.
  • In the architecture you described, what was the biggest risk you underestimated, and how did it actually surface once you were in production?
  • Was there a time you disagreed with a decision but committed to it anyway? How did it turn out, and would you make the same call again?
  • For the ownership story you shared, how did you balance taking on that extra responsibility against your existing commitments? Did anything else slip as a result?
  • Describe a major architectural decision you made that turned out to be wrong. How did you find out, and how did you walk it back?

“This is the one I spent the most prep time on and still felt like I left points on the table.”

View Post

Uber·Software Engineer·Technical Phone Screen

Jun 2026
Uber SWE interview with a grid simulation problem that looked deceptively straightforward but had a lot of edge cases hiding underneath. The kind of question where you think you're done and then they ask about holes in the pattern and you realize you missed half the problem.
  • You have an n-by-m grid and a list of binary tile patterns. For each pattern in order, scan the grid row by row to find the first valid top-left position where the pattern fits without going out of bounds and without overlapping already-filled cells. If found, place it and mark those cells with the pattern's index (1-based). If not, skip it. Output the final grid. Discuss scanning order, tie-breaking, complexity, and edge cases like holey patterns or patterns bigger than the grid.

“I spent the first few minutes just restating the problem back to make sure I had the scanning order right, which actually helped because I initially assumed you'd place greedily without caring about holes.”

View Post

Uber·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Did a coding round for a Software Engineer role at Uber, four algorithmic problems back to back. Nothing too wild in terms of topics but the combination lock one tripped me up more than I expected.
  • You have floors 1 through N and an API that tells you whether a given floor is safe for elevator operation. There's a threshold floor T where everything at or below is safe and everything above is not. Find T using only O(log N) API calls.
  • A combination lock has four wheels, each ranging from 0 to 9 with wraparound. Each move changes exactly one wheel by one step. Given a start state, a target state, and a list of blocked states you cannot pass through, return the minimum number of moves to reach the target or -1 if it's impossible.
  • Given a 2D grid with walls and open cells, and a starting position inside the grid, find the minimum number of moves to reach any open cell on the boundary of the grid that isn't the starting cell. Return -1 if no such exit is reachable.
  • Generate a minesweeper board: given dimensions and a mine count N, randomly place N mines in distinct cells, then fill every non-mine cell with the count of mines in its eight neighboring cells. Focus on clean, simple code without unnecessary extra data structures.

“Pretty transparent binary search setup once you see it.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Uber SWE interview that mixed behavioral questions with a live system design session. You walk through a project you led, then they put you on the spot to whiteboard the architecture right there, defending every call you made. Felt more like a senior-level pressure test than a standard loop.
  • Walk me through a recent project you led end-to-end. What was your role, what decisions did you own, and what was the measurable impact?
  • Now whiteboard the system architecture for that project. Show me the components, how data flows through them, and where you'd make different scaling decisions if the load grew 10x.

“The behavioral part felt like a warm-up but it wasn't.”

View Post

Uber·Software Engineer·Online Assessment (OA)

Jun 2026
Uber SWE online assessment on HackerRank, camera on the whole time which was a little uncomfortable. One hard DP problem, took a while to crack it but got there eventually.
  • Given an array of integers, find the maximum sum you can achieve by jumping through elements where each jump length must be a prime number (a variant of the jump game DP problem).

“Took me longer than I'd like to admit to even figure out what the prime condition was changing about the state transitions.”

View Post

Uber·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Uber system design round focused entirely on a real-time Top-K ranking system, the kind of thing you'd use for driver leaderboards or restaurant rankings. Four parts back to back: core data structures, concurrency, batch ingestion with error handling, and testing. Pretty dense for one session.
  • Design an object-oriented, real-time Top-K ranking system that supports update, top_k, and remove operations on a large set of entities with numeric scores.
  • How does your design change when multiple threads are calling update, remove, and top_k concurrently? How do you keep the two structures consistent and scale write throughput?
  • How would you handle very large input batches where some records are malformed or fail during processing, without letting bad records abort the whole batch?
  • What tests would you write to verify correctness, edge cases, and concurrency for this system?

“The core trap I nearly fell into was reaching for a single heap and calling it done.”

View Post

Uber·Software Engineer·Online Assessment (OA)

Jun 2026
Took an OA for a Software Engineer role at Uber and got tripped up on a string/palindrome problem. Couldn't crack the optimal solution under time pressure and TLE'd on some cases, so now I'm just waiting and hoping it wasn't a dealbreaker.
  • Given a string, compute the total cost of converting every substring into a palindrome. For example, with the string "abc", the substrings "ab", "bc", and "abc" each cost 1, giving a total of 3.

“I got a working solution but it was nowhere near O(n) and it showed.”

View Post

Uber·Data Scientist·Onsite - Product Sense / Strategy

Jun 2026
Uber DS onsite with a deep metrics design question for a rides marketplace. The whole thing was one long problem and they kept pushing for more rigor than I expected.
  • Design a suite of metrics that captures rider experience in a rides marketplace purely from a supply-demand perspective, excluding app UI and vehicle quality. Then propose a single composite index with an explicit formula and weights.
  • How would you define censoring rules for unfulfilled ride requests, and how do you handle exclusions like holidays or platform outages in your metric calculations?
  • Describe how you would backtest your proposed metric suite to show it responds to supply and demand shocks but stays stable during UI-only changes.

“This one took me a while to even understand what they were really asking.”

View Post