Skip to content
Uber Coding Interview Questions: Patterns and Top Problems

Uber Coding Interview Questions: Patterns and Top Problems

Every company page on this site compares difficulty, and Uber stands apart. More than a third of its tagged problems are Hard, against roughly one sixth everywhere else. The list is also graph-soaked. Bus Routes, Alien Dictionary, and two versions of Number of Islands sit in the top band, which mirrors a business built on routes, maps, and matching.

This page ranks the questions that appear most often in a recent tagging window for Uber. Each row links to the pattern guide on this site that teaches the technique. The later sections break down the topic weights and build a study order for the hardest bank on the site.

The short version

  • Hard problems make up 37 percent of the set. Only 7 percent rank Easy.
  • Union find holds about 16 percent of tags. No other software giant we cover comes close.
  • Arrays touch about 63 percent of tagged problems. Graph rows reward connectivity skill.
Every question below exists as a review card in our app, grouped by pattern. Build your Uber deck and train on methods rather than solutions.

Uber’s Interview Loop

  • A recruiter call covers your background and target level.
  • One technical phone screen follows, usually in CoderPad with runnable code.
  • The onsite loop runs about four rounds.
  • Two rounds are coding. Expect practical framing around maps, matching, or rates rather than abstract puzzles.
  • One round covers system design for most seniority levels.
  • A behavioral round checks collaboration, and candidate reports consistently mention a real bar here.

Two details shape your prep. First, code must run in the phone screen, so test as you write instead of promising it compiles. Second, the coding rounds favor problems that look like Uber’s domain. Practicing route, interval, and connectivity flavored problems pays directly.

The Questions Uber Asks Most

LeetCode tags problems by company from candidate reports and scores each tag on a relative frequency scale from 0 to 100. These 20 score highest across a recent tagging window for Uber. Each row links to our guide for the technique behind it.

If you learn five guides first, learn these

  1. Start with graph traversal . It covers Bus Routes and Number of Islands in the top band.
  2. Pair it with union find . Two top rows test connectivity directly.
  3. Study binary search next. Three top rows use it in different forms.
  4. Add monotonic stack for the subarray limit row. It also covers Final Prices.
  5. Finish with topological sort . Alien Dictionary rewards clean ordering logic.
ProblemDifficultyDemandPattern Guide
First Unique Number
Medium
Very high
Queue
Minimum Edge Reversals So Every Node Is Reachable
Hard
Very high
Dynamic Programming
Bus Routes
Hard
Very high
Graph Traversal
Number of Islands II
Hard
Very high
Union Find
Longest Continuous Subarray With Absolute Diff Less Than or Equal to Limit
Medium
Very high
Monotonic Stack
Construct Quad Tree
Medium
Very high
Tree Traversal
Alien Dictionary
Hard
Very high
Topological Sort
Count Paths That Can Form a Palindrome in a Tree
Hard
Very high
Bit Manipulation
The Earliest Moment When Everyone Become Friends
Medium
Very high
Union Find
Squares of a Sorted Array
Easy
Very high
Two Pointers
Final Prices With a Special Discount in a Shop
Easy
Very high
Monotonic Stack
Word Search
Medium
Very high
Backtracking
Kth Smallest Element in a BST
Medium
Very high
Tree Traversal
Leftmost Column with at Least a One
Medium
Very high
Binary Search
Number of Islands
Medium
High
Graph Traversal
Time Based Key-Value Store
Medium
High
Binary Search
Find Peak Element
Medium
High
Binary Search
Minimum Number of Refueling Stops
Hard
High
Heap Priority Queue
Design Hit Counter
Medium
High
Queue
Insert Delete GetRandom O(1)
Medium
High
Hash Table

Look at how few famous Easy warmups appear here. Two Sum is absent from the top band entirely, and Squares of a Sorted Array is the only Easy near the very top. Where Google screens basics, Uber starts closer to working difficulty and pushes into graph territory fast.

Hard problems punish fuzzy memory worst. Queue this list as spaced repetition cards so every graph template stays assembled under pressure.

Topics and Difficulty at Uber

The most common topics among tagged Uber problems appear below. Shares overlap because topics repeat inside one problem.

TopicTagged ProblemsShare
Array Manipulation36~63%
Graph Traversal (BFS)17~30%
Graph Traversal (DFS)15~26%
Sorting12~21%
Hash Table11~19%
String Manipulation11~19%
Matrix10~18%
Design9~16%
Union Find9~16%
Heap Priority Queue9~16%
Binary Search9~16%

Sorting carries no link because we treat it as part of other patterns rather than a pattern of its own, and design has no single guide because it spans several. Two numbers have no parallel elsewhere on this site. Breadth-first and depth-first search combine to touch more than half of all tagged problems when you count their overlap. And union find holds 16 percent of tags against 4 percent or less at every software giant we cover. Connectivity questions are an Uber signature.

Difficulty confirms the reputation:

DifficultyProblemsShare
Easy
47%
Medium
3256%
Hard
2137%

Only four Easy problems appear in the whole recent window. Apple already skips warmups at 10 percent Easy, and Uber goes further at 7 percent. Meanwhile Hard triples the big-tech norm. Plan for a loop where the first question already needs a full template executed cleanly.

Method: problems carrying the Uber tag on LeetCode in a recent tagging window, ranked by relative frequency scores from 0 to 100. Shares overlap because one problem can carry several topics.

Where to Spend Your Prep Time

  1. Make graphs your first language. BFS, DFS, and union find together dominate this list like nowhere else. Work the graph traversal and union find guides until grid and connectivity problems feel routine.
  2. Learn topological sort properly. Alien Dictionary is one of the most respected Hard problems anywhere, and dependencies show up across Uber’s rounds. The topological sort guide covers the ordering logic.
  3. Add monotonic structures for sliding range problems. The long-titled subarray limit problem and Final Prices both need them. The monotonic stack guide teaches the core pattern.
  4. Practice stating domain assumptions out loud. Uber interviewers frame problems as routes, drivers, or events. Restate the model you chose and why before coding, since that translation step is part of the signal.
  5. Review on a schedule with extra spacing for Hards. Hard templates decay fastest under pressure. Spaced repetition returns each one right before you would lose it.
    graph TD
    A["Graph traversal basics"] --> B["Union find connectivity"]
    B --> C["Topological ordering"]
    C --> D["Monotonic stack patterns"]
    D --> E["Binary search forms"]
    E --> F["Spaced repetition for Hard templates"]
  

The diagram orders the prep plan from graph skill through ordering and range patterns to steady review.

A hard bank makes steady review the only sane strategy. Start your Uber review schedule and let spaced repetition hold the templates together.

Questions Candidates Ask

Why is Uber called the toughest bank here?

Hard problems make up 37 percent of the set. Only 7 percent rank Easy, with four Easy problems in the whole window. Expect the first question to need a full template run cleanly.

What topic defines Uber interviews?

Breadth-first search touches about 30 percent and depth-first about 26 percent. Union find holds about 16 percent, far above the share at other companies. The union find guide covers Number of Islands II and the Earliest Moment row.

Where should I start?

With the graph traversal guide. Pair it with union find until grid and connectivity problems feel routine. Then take binary search for its three top rows.

Next Steps

Work the table top to bottom, expecting the graph rows to take longest. The coding patterns index lists every guide, and graph traversal plus union find are the two reads that matter most for Uber. To see the opposite end of the difficulty spectrum, compare the Meta speed list and the Stripe practical set.

You know the hardest bank on the site now. Turn it into scheduled reviews so every graph template survives contact with the real loop.