Software Engineer · Phone Screen · Coding

Uber — Software Engineer ❌ Failed

Level: Senior-Level

Round: Phone Screen · Type: Coding · Difficulty: 7/10 · Duration: 60 min · Interviewer: Unfriendly

Topics: Arrays, Graph Traversal, Matrix Traversal

Location: San Francisco, CA

Interview date: 2025-12-31

Got offer: False

Summary

I had a technical phone screen with a good interviewer. I didn't fully understand the question initially and rushed into coding, which led to insufficient time for optimization. I ultimately failed the round.

Details

The question I received was:

You are given an m x n integer matrix terrain, and an array limits of size k. You must compute an array answer of size k such that for each limits[i]:

1. Start at the top-left cell (0,0).
2. Repeat:
   • If limits[i] is strictly greater than the value of the current cell, then:
     • You gain 1 point the first time you visit this cell.
     • You may move to any adjacent cell (up, down, left, right).
   • Otherwise:
     • You gain no points.
   • The process ends.
3. After the process, answer[i] is the maximum points achievable.

You are allowed to revisit cells multiple times, but points are only earned on the first visit. Return answer. Traversal ends once we meet a value in the grid that is >= than current limit.

Input: int[][] matrix, int[] limits Output: int[] answer

Example:

{1,4,2,8} {0,4,0,8} {1,2,0,8}

limit = 8 => 9 limit = 2 => 3

{1,2,3}, 1 => [2,2] {2,5,7}, {3,5,1} {5,6,2}

result [5,8,1]

My approach and key insights:

I failed to correctly understand the problem statement initially, and I started coding prematurely. The problem involves graph/matrix traversal with certain constraints. The core idea is to explore the matrix while the value in current cell is less than given limit. This requires a backtracking or DFS approach. A key detail is to only count points on the first visit.

Preparation Tips:

• Practice graph traversal and matrix traversal problems.
• Focus on understanding problem constraints and edge cases thoroughly before starting to code.
• Improve time management during coding interviews.