Google interview notes, page 204 of 215

FAANG / Big Tech

Most of these Google write-ups come from Software Engineers (1,167), then Product Managers (686) and Data Scientists (125). The most reported rounds are the product sense round (667), the technical phone screen (660), the behavioral / leadership round (252) and the onsite coding round (214).

2146 interview notes · updated Oct 2026

Google·Software Engineer·Onsite - Behavioral / Leadership

Jul 2026

Behavioral round at Google for a software engineer role. Pretty standard stuff but a couple questions had enough nuance to make me think twice, especially the ownership one with the two variants.

Tell me about a time you disagreed with a peer, cross-functional partner, or manager. How do you generally handle conflict?
Describe a situation where different stakeholders had conflicting perspectives that shaped the design direction. How did you bring those views together into a better final outcome?
What do you do when a deadline is extremely tight? How do you manage multiple deadlines competing at the same time?
You're about to start a new project tomorrow and you discover a bug in another team's codebase that you don't own. What do you do?
Same scenario, but the bug is in a project you own. The two projects have no direct dependencies. Does your approach change, and why?
In what ways do you think a large company like Google could do better?

“I had an example ready but I kept second-guessing whether the conflict was 'big enough' to sound meaningful.”

View Post

Google·Product Manager·Onsite - Product Sense / Strategy

Jul 2026

Google PM interview with a classic estimation question about airport metal detector throughput. Pretty standard case format but the specificity of the scenario tripped me up a bit.

Pick an airport and estimate how many people pass through the metal detectors in a single day.

“I picked a mid-size airport thinking it would be easier to reason about than JFK or LAX, which in hindsight was probably the right call.”

View Post

Google·Software Engineer·Hiring Manager Screen

Jul 2026

Interviewed at Google and got asked the classic 'tell us what you've built that's relevant here' question. Pretty standard screen, nothing too wild.

Which projects from your past work do you think are most relevant to what we do here?

“I rambled a bit trying to cover too many things at once.”

View Post

Google·Product Manager·Onsite - Product Sense / Strategy

Jul 2026

Google PM interview, got hit with a pretty open-ended Gen AI product design question. Not a lot of structure to the original prompt so I'm going off memory here, but it was the kind of question that sounds breezy until you're actually in it.

How would you approach building a product that leverages generative AI?

“I went straight to user problems first, which felt right, but I spent too long on that part and rushed the actual Gen AI integration angle.”

View Post

Google·Product Manager·Onsite - Behavioral / Leadership

Jul 2026

Interviewed for a PM role at Google and got a question I wasn't expecting to see in a product interview at all. It was more about communication and interpersonal polish than product thinking, which threw me off a bit.

In a remote meeting, if you can't make out someone's name or question because of a heavy accent or bad audio, what do you actually say and do to clarify without derailing the conversation?

“I fumbled this more than I'd like to admit.”

View Post

Google·Software Engineer·Hiring Manager Screen

Jul 2026

Went through what seemed like a leadership-focused screen at Google, and this question about handling non-team-players was pretty much the centerpiece of it.

How do you handle someone on your team who isn't being a team player?

“I went straight into a story about a direct conversation I had with someone who kept siloing their work.”

View Post

Google·Software Engineer·Technical Phone Screen

Jul 2026

Google SWE interview with a grid-based constraint satisfaction problem. The problem looked like a straightforward placement puzzle but the house adjacency rule on top of the row/column uniqueness made it a real backtracking exercise. No outcome info shared.

Given an N x N grid with houses ('H') and empty cells ('0'), place flowers ('F') such that each row and each column contains exactly one flower, and every house has exactly one flower among its 4-directional neighbors. Return a valid grid or indicate no solution exists.

“The row/column uniqueness part felt like N-queens to me immediately, so I went straight to backtracking.”

View Post

Google·Data Scientist·Technical Phone Screen

Jul 2026

Google data scientist interview that went deep on imbalanced classification, specifically a multi-part question covering metric selection, threshold analysis in Python, AUPRC computation, and cost-sensitive threshold picking. The whole thing felt like one long problem that kept branching.

You have a dataset with true labels and predicted probabilities, and the positive class is only about 5% of the data. Which evaluation metrics would you prioritize and why? What are the pitfalls of using accuracy or ROC-AUC in this kind of heavy class imbalance?
Write a Python function that takes the actual labels and predicted probabilities and returns thresholds, precision, recall, and F1 across all unique predicted probability values. Handle ties in scores, empty denominators, and enforce monotonic precision if needed.
How would you compute AUPRC efficiently, and what effect does score calibration have on the PR curve?
How would you choose an operating threshold given that false positives and false negatives have different costs, and there are volume constraints on how many positives you can act on?

“This part I actually felt okay about.”

View Post

Google·Data Scientist·Technical Phone Screen

Jul 2026

Google stats/ML interview, just one question but it was a conceptual one that required more nuance than I expected going in.

Why might a biased estimator sometimes be preferable to an unbiased one?

“I knew the textbook answer about variance and the bias-variance tradeoff, but I fumbled explaining when you'd actually make that call in practice.”

View Post

Google·Software Engineer·Onsite - Multi Round

Jul 2026

Interviewed for a software engineering role at Google's Warsaw office, four rounds total: two remote and two in-person. Passed the first two but got knocked out by a dynamic programming problem in the onsites, which put me on a year-long cooldown. The experience was a reality check about how little Google cares about anything except whether you can produce an optimal solution under pressure.

Given a 2D grid, search for a target word by traversing adjacent cells using depth-first search.
Given a set of coin denominations and a target amount, find the number of distinct combinations that sum to that target.
Given the previous coin change problem, can you reverse the problem: given a number of combinations, reconstruct a valid set of coin denominations?
Partition an array into subarrays satisfying a given condition, using prefix sums to optimize the solution.
Behavioral questions covering past work situations, including at least one hypothetical scenario about how you'd handle a specific challenge.

“Solved it, but missed a space optimization that I only realized after the fact.”

View Post