Updated by Google candidates
Software engineer intern Interview Experience
There was no behavioral portion at all. Introductions took less than a minute, then we went straight into the problem.
Interview process
I applied online, sent the recruiter my transcript, graduation date, and availability, and then got an online assessment before they scheduled me. After that I had two 45-minute coding interviews in a scratchpad. There was basically no behavioral portion at all, and even the intros were super short before we jumped into the problem. The questions themselves were classic algorithm problems: one shortest-path style graph question and one meeting rooms question with an overnight twist. It felt very stripped down and technical the whole way through.
- Recruiter screen
- Online assessment
- Technical interview
Interview rounds
3 questions across 3 roundsRecruiter screen
1 questionOverview
I applied online, and the recruiter conversation was basically just logistics before they moved me forward.
Specific questions asked
Can you send your transcript, graduation date, and interview availability?
I sent over my transcript, my graduation date, and the times I was available to interview. It felt pretty straightforward and administrative, not evaluative.
Online assessment
Overview
After I sent the recruiter info, I got an online assessment before they let me schedule the live rounds.
Question types asked
Technical interview
2 questionsOverview
I had two 45-minute coding interviews in a scratchpad that felt like a Google Doc. There was basically no behavioral part at all, and the intro was under a minute before we jumped straight into coding. It was very technical and detail-oriented, and they could give a follow-up if you finished early.
Question types asked
Specific questions asked
Solve this shortest-path style graph traversal problem.
I approached it like a classic graph traversal and shortest paths question. I focused on getting to the most optimal time and space complexity instead of just any working solution. The interviewer was really watching the details, so beyond the algorithm itself I had to make sure the code was clean and correct all the way through.
Find the minimum number of meeting rooms needed to host overlapping meetings.
How would you handle meetings that cross into the next day, like 11:30 PM to 1:30 AM?
I solved it like the standard meeting rooms problem first, then had to handle the overnight edge case where a meeting could start late one day and end after midnight. That made it trickier than the usual version. The interviewer also cared a lot about code quality, so they called out off-by-one mistakes and pushed me to clean them up before ending the round.
Interview tips
I would practice getting to the optimal time and space complexity quickly, because they really do care about that and not just whether you can get something working. I would also leave myself time to debug carefully, especially off-by-one mistakes, because they were looking closely at code quality. And I would prep edge cases, since even a standard problem can turn into something harder with one twist, like meetings crossing midnight. If you finish early, I would expect them to keep going and give you more.
Company culture
Going through it, I felt like they care a lot about technical excellence and also about whether I can communicate clearly while coding. The interviewer style felt very detail-oriented. They were watching closely and pointing me toward mistakes, especially small correctness issues, instead of just letting sloppy code slide.
Get full access with a membership, or share your experience to try it free.
More Google interview experiences
“It was basically the most by-the-book Google loop possible: shortest path in a DAG, minimum meeting rooms, then a news feed design where they kept pushing on fanout, storage, and ranking.”
“The system design wasn’t generic. It was built around the team’s actual work: an AI agent that lets enterprise users search and query their data and documents.”
