Google L4 Software Engineer Interview, India

GoogleL4 Software Engineer·India·Interviewed November 2024Offer

Updated July 17, 2026

The candidate was a backend engineer with more than five years of experience, working at a large tech company at the time of this interview. An earlier attempt to interview at Google, several years prior, had ended at the phone screen after only three weeks of preparation. This time, the candidate set aside close to a year for dedicated data-structures-and-algorithms preparation before applying again for an L4 backend role in India.

The loop consisted of a phone screen, three onsite rounds focused on data structures and algorithms, and a separate Googliness round covering behavioral topics and a resume deep dive. After the onsite rounds, the process continued through team matching, a Hiring Committee review, and a compensation discussion before an offer was extended.

How the process went

Preparation
Roughly a year of focused practice on medium and hard problems, self-recorded mock interviews, and targeted study of topics such as disjoint set union, segment trees, and dynamic programming with space optimization.
Phone screen
One round with a graph-traversal problem and several follow-ups; feedback came back about two weeks later, and the candidate took roughly a month to prepare further before the onsite.
Onsite
Three back-to-back data-structures-and-algorithms rounds, each about 45 minutes, conducted through a shared document or a plain online text editor with no code execution.
Googliness round
A separate round covering a deep dive into a personal project from the resume and standard behavioral questions.
Team match
Positive feedback was shared a few days after the onsite; the team-matching process itself took about 40 days, including one team that dropped out a day before a scheduled call.
Hiring Committee and offer
The profile was approved at the L4 level about a week after the team match, followed by a compensation discussion and an offer two days later.

Phone Screen

Graph traversal · 45 minutes

A problem centered on multi-source BFS, applying graph traversal to analyze relationships within a dataset
Follow-up on handling an edge case involving elements with unique constraints
Follow-up on extracting additional insights from the computed results
Follow-up asking for a different algorithmic strategy once the problem constraints were changed

Recruiter feedback took about two weeks and was described as clearly positive, with an invitation to move to onsite rounds.

Onsite Round 1

Graph traversal (BFS vs. DFS)

A graph problem where choosing correctly between BFS and DFS was central to the solution
Follow-up on finding the longest cycle in an undirected graph

Onsite Round 2

Queues, with system-design elements mixed in

A problem combining data-structure work with system-design considerations, centered on queues and patterns similar to handling asynchronous calls

Onsite Round 3

Connected components

A problem requiring identification of connected components within a structured dataset, where the candidate had to clarify the input format before solving
Follow-up on counting the number of distinct components
Follow-up on determining the size of each component
Follow-up on identifying patterns among the components

The interviewer asked the candidate to come up with original test cases rather than relying only on the ones provided.

Googliness Round

Behavioral discussion and resume deep dive

A roughly 30-minute deep dive into a personal project listed on the resume, with the conversation focused on UI/UX design choices even though the candidate had applied for a backend role
How do you handle conflicts in a team?
How would you mentor a new intern?

The interviewer initially mistook this for a technical round and began with a DSA problem before realizing it was the Googliness round and switching to the resume discussion.

Key takeaways

Before writing code, state your approach out loud along with its time and space complexity, then dry-run it against your own test cases rather than only the ones the interviewer gives you.
Recording mock interviews of yourself (screen and webcam) can surface gaps in how clearly you explain your reasoning, separate from whether the solution itself is correct.
For graph-heavy loops, build comfort with BFS versus DFS trade-offs, connected components, and cycle detection, since a single base problem can carry several follow-ups in that area.
Be ready to discuss any project on your resume in depth, even if the conversation drifts into areas outside your target role's core skills.
Once team matching starts, expect the process to stretch over weeks; a positive onsite result did not by itself determine which team or level the candidate would land at.

Source

The questions and process facts come from the candidate's public write-up, linked below. The retelling above is our own summary.

Candidate's public write-up on LeetCode Discuss