Google SDEIII L4 Interview Experience
I recently went through the interview process for an SDE3 L4 role at Google.
There were four rounds in total — two virtual rounds followed by two back-to-back onsite rounds.
Round 1 (Phone Screening):
This was a DSA round. The problem was based on arrays — given an array, we had to check whether it could be divided into groups of size 5 such that each group contains consecutive increasing numbers. The output was a boolean (true/false).
The discussion mainly revolved around handling duplicates properly and ensuring all elements are used exactly once.
Round 2 (Behavioral / Team Fit):
This round was more conversational. Questions were around past experiences — times when I took ownership, made mistakes, handled disagreements, etc.
One situational question was about planning a team outing where people have different preferences and are not aligned — how to handle that and reach a decision.
Overall positive feedback so moved to onsite interviews.
Round 3 (Onsite – DSA):
I was given a set of points on a 2D plane (array of [x, y] pairs). The task was to find the maximum area of a rectangle that can be formed using those points.
The challenge here was not just identifying rectangles but doing it efficiently. I discussed multiple approaches and reached an optimized solution, but due to time constraints, I couldn’t fully complete the code. Most of the time went into reasoning about the geometry and optimizing the approach.
Round 4 (Onsite – DSA):
This round was based on task scheduling.
First part: Given multiple tasks (all with equal execution time) and a set of machines, determine the earliest time by which all tasks can be completed.
Second part: Given the number of tasks and execution time, find the minimum number of CPUs required to achieve that earliest completion time.
The key idea here involved combining scheduling logic with binary search to optimize resource usage.
Overall, the interviews were heavily focused on problem-solving, optimization, and clarity of thought. Communication played a big role, especially in explaining approaches and trade-offs.
Verdict: Not positive
Cool-off period: 12 months