Uber SWE Interview: Coding Interview Guide

Estimated read time: 8-10 minutes

Summary: The Uber SWE coding interview in the loop checks repeatable coding ability across one or more rounds. Expect data structures, algorithms, optimization, testing, and changed constraints.

At a glance

• Stage: Loop.
• Round: Coding interview.
• Typical duration: 45-60 minutes each when reported.
• Likely interviewers: software engineers.
• Relevant levels: intern through staff-plus, possible or role-dependent.

What happens in this round

Loop coding rounds go deeper than the screen. Public reports mention trees, graphs, intervals, scheduling, heaps, top-k problems, rate-limiter-like variants, and debugging or optimizing initial code.

Expect follow-ups. Uber's marketplace and routing context can make constraints around geography, timing, matching, and availability feel natural, but exact questions are not reliably public.

Level-specific expectations

Intern and new grad candidates should show fundamentals and recover quickly from mistakes.

Junior and mid-level candidates should show robust implementation and testing.

Senior and staff candidates should show coaching-level clarity, maintainability, and systems-adjacent tradeoffs.

Candidate-facing questions to prepare

• Given interval schedules for drivers, find available windows and handle overlapping trips.
• Use a heap to return the top-k nearby or highest-priority items from a stream.
• Traverse a tree or graph and return a result under time or distance constraints.
• Implement a simple rate limiter and explain how the design changes in a distributed setting.
• Debug initial code that fails on disconnected graphs, duplicates, or boundary timestamps.
• Optimize a working solution when input size or latency requirements increase.
• Explain what tests prove correctness and which risks remain.

Strong signals

• Stable problem decomposition across multiple rounds.
• Correct code with edge-case tests.
• Comfort changing constraints.
• Clear tradeoffs around latency, memory, and correctness.
• Senior-level communication and maintainability awareness.

Common failure modes

Overcomplicating early. Build the smallest correct version first.

Not validating assumptions. Marketplace-style data often hides edge cases.

Weak testing. Unresolved bugs can dominate a loop round.

How to prepare

• Practice graphs, intervals, scheduling, heaps, top-k, and rate limiting.
• Use Uber-like data only as a framing device, not as a guarantee.
• Test edge cases and explain complexity.
• For senior roles, discuss maintainability and design-adjacent follow-ups.
• Ask whether the loop has multiple coding rounds.