Google PhD New Grad L4 Europe Onsite: Tree Templates, Top-K IPs, and Skill Matching

Role: Software Engineer, PhD New Grad — L4 Europe
Result: Waitlisted during Team Match
Report time: July 2026

Interview style

Most interviewers were based in the United Kingdom or Poland. The candidate found the questions manageable and felt that clear communication mattered more than raw difficulty. The behavioral round used familiar questions without unusual surprises.

Coding round 1: template across tree leaves

Each leaf of a binary tree contains a string, character, or space. Concatenate the leaf contents into one long string and determine whether a given template occurs in it. The tree was framed as being generated from an HTML file.

The candidate used brute-force matching, and the interviewer accepted it. They also analyzed complexity.

Practice: Find a Template Across Binary-Tree Leaves

Coding round 2

Generate strings from 1 and 2

Output every string made from 1 and 2 whose values sum to N, then give the exact time complexity.

Practice: Generate One-Two Strings With Sum N

Top-K IP addresses from a large file

Given a gigabyte-scale file of IP addresses, return the k most frequent IPs and analyze exact time and space complexity.

Practice: Top-K IP Addresses From File Records

Coding round 3: programmer-problem matching

There are N problems and M programmers. Each problem has tags, and each programmer has skills. A programmer may receive a problem when at least one skill intersects with one tag. Each programmer and problem can be assigned at most once.

The base question asked for any feasible assignment. The candidate used DFS as a greedy approach and explained that even one matched pair would satisfy the base requirement.

Follow-ups asked how to:

1. Maximize the number of matches using bipartite matching.
2. Design the capability in a distributed environment.
3. Support streaming data.

The candidate discussed the follow-ups without implementing them.

Practice: Maximum Programmer-Problem Matching

Design practice: Distributed Streaming Skill Matching

The candidate said European Team Match was particularly difficult and invited others in the same situation to compare notes.