Google L3 Software Engineer Interview Experience — Bangalore, 2020

GoogleL3 Software Engineer·Bangalore, India·Interviewed August 2020Offer

Updated July 17, 2026

This account comes from a candidate who interviewed for an L3 Software Engineer role at Google's Bangalore office in 2020. The candidate first applied to two positions in March 2020 through a friend's referral, but HR replied within two days saying their roughly two years of experience didn't match those postings. Months later, a Google recruiter reached out directly after finding the candidate's resume already in the company's system, saying a hiring manager had reviewed it and wanted to move forward with an interview.

The candidate already held a competing offer, so the recruiter compressed the usual timeline: outreach began in mid-July, and the phone screen plus five virtual onsite interviews were completed within about three weeks, followed by an offer in mid-August. The candidate later wrote up a round-by-round account along with their preparation approach, which is summarized below.

How the process went

Application (referral)
Applied to two roles via a friend's referral in March 2020; HR replied within two days saying the candidate's roughly two years of experience didn't match the postings.
Recruiter outreach
In mid-July 2020, a Google recruiter reached out after finding the candidate's resume already in Google's system; a hiring manager had reviewed it and wanted to set up an interview.
Phone screen
A virtual phone screen took place in the last week of July, covering coding questions and a discussion of a difficult problem from work.
Onsite loop
Following positive phone-screen feedback, five virtual onsite interviews were scheduled across three days in the first week of August.
Hiring committee and team matching
In the second week of August, the packet went through hiring-committee review and the candidate was matched with a team.
Hiring manager calls and offer
The recruiter arranged calls with a couple of hiring managers to discuss team fit; an offer was extended in mid-August, a few days after the candidate found a team they were excited about.

Phone Screen

Coding (virtual)

Count the number of distinct events that occurred within a fixed time window (approached with a hash map).
Describe a difficult problem you solved and how you approached it (candidate discussed a recent work issue).
Given an Excel-style column number, return the corresponding column label (base-26 number system).

Recruiter feedback described this round as strong, citing coding speed and clear solution formulation.

Onsite 1

Coding — arrays and dynamic programming

An array problem solved using prefix and suffix sums.
A follow-up variation of the same problem solved with dynamic programming.

Candidate solved both problems; an initially suboptimal space complexity was improved after a nudge from the interviewer, and a missed boundary condition was caught while dry-running an example.

Onsite 2

Behavioral ("Googleyness")

General work-scenario questions covering leadership, a conflict with a colleague, and delivering under pressure.
A technical discussion about testing a new feature in Google Photos.
A hypothetical scenario about how to handle team members unwilling to join a team-bonding activity.

Candidate described this round as going well overall.

Onsite 3

Coding — trees and binary search

Find the last node in level order of a complete binary tree, using the parent/child index relationship (parent n maps to children 2n+1 and 2n+2).
A follow-up applying binary search on top of the function written for the first part.

Candidate described this as the weakest of the five onsite rounds: it took a while to reach the first solution even with hints, and the second part wasn't fully coded due to time constraints. The interviewer was described as arriving late and less engaged than others in the loop.

Onsite 4

Coding — heaps and binary search

Find the kth largest element in a running stream of numbers, with k fixed (a heap-based approach).
A follow-up where the value range in the stream is smaller but k is variable, solved with binary search over a prefix sum of value frequencies.

Candidate reported a strong round, including a discussion of overhead reductions for heap push/pop operations; a missed boundary condition was corrected after the interviewer pointed it out.

Onsite 5

Coding — recursion and graphs

Find the farthest leaf from the root in a weighted n-ary tree, including designing the data structure for the tree (recursion).
The same problem extended to a graph instead of a tree, recognized as equivalent to finding the node with the maximum value in Dijkstra's distance array.

Candidate described this as a near-perfect round per recruiter feedback, citing a clear thought process and bug-free code.

Hiring Manager Calls

Team matching

Domain- and behavior-related questions from a couple of hiring managers evaluating team fit.

Candidate noted these calls are informal in name but can still involve substantial questioning, and used them to ask about the team's work and technologies.

Key takeaways

Favor depth over volume in coding practice — the candidate solved around 260 problems with roughly a 25/50/25 easy/medium/hard split rather than grinding a much larger count.
Practice narrating your approach out loud while solving problems; the candidate found that systematic, incremental thinking made it easier to communicate with interviewers and avoid silent stretches.
Prepare at least two stories per common behavioral scenario so you have backups when scenarios overlap, and be ready to discuss genuine strengths and weaknesses rather than rehearsed answers.
Focus preparation on frequently tested areas — dynamic programming, trees, graphs, binary search, sliding window, and basic combinatorics — rather than less common topics like advanced geometry algorithms.
Treat hiring-manager calls as a two-way conversation: use them to ask about team scope and technologies, since accepting an offer also means committing to a specific team.

Practice a Google interview

Rehearse out loud against the kinds of questions in this story — with an AI interviewer that asks follow-ups.

Practice this interview

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