← Back to Directory

Google interview notes, page 160 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·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Google PM interview, one question about how you'd orient yourself in a new role and figure out what to actually work on. Pretty open-ended, which I did not love.
  • You've just started as a PM at a new company. How do you figure out which problems are worth working on?

“I fumbled the opening by jumping straight into tactics (talk to users, look at metrics) before establishing any kind of structure.”

View Post

Google·Software Engineer·Recruiter / HR Screen

Jun 2026
Applied to Google as a Software Engineer through external recruiters, made it to a 17-minute screening call where the recruiter said the level and team weren't even decided yet, then got rejected the same day. The aftermath was somehow weirder: the recruiter offered to set up a follow-up call, then two hours later sent a canned final rejection email.
  • How have you approached latency reduction and performance optimization in your backend work?
  • What is your preferred balance between backend and frontend work?

“This is where I fumbled.”

View Post

Google·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Google PM interview with a single product design question that sounds deceptively simple but has a lot of moving parts once you actually dig in. The 80% market capture constraint made it feel more like a strategy case than a pure ideation exercise.
  • Design a new communication device for elderly users that could capture 80% of the market, with smartphones explicitly off the table.

“I spent the first couple minutes just trying to figure out what 'capturing 80%' actually meant as a constraint.”

View Post

Google·Software Engineer·Technical Phone Screen

Jun 2026
Got a product analytics question at Google focused on diagnosing a metric drop in Shopping. Pretty standard diagnostic case but the specificity of the placement context tripped me up a bit.
  • How would you investigate a 3% drop in CTR on Google Shopping placement?

“I started with the usual segmentation angle: device, geography, query type, time of day.”

View Post

Google·Machine Learning Engineer·Onsite - Behavioral / Leadership

Jun 2026
Behavioral round for an ML Engineer role at Google, roughly an hour with some small talk at the start. Pretty standard stuff but the prep list they gave me was actually useful.
  • Tell me about a project you worked on that had a meaningful, measurable impact.
  • Describe a time you disagreed with a teammate or manager. How did you handle it?
  • Walk me through a failure or significant mistake and what you took away from it.
  • How have you dealt with shifting priorities or requirements that weren't clearly defined?
  • Why do you want to work on this specific team at this company?

“This is the one you absolutely cannot wing.”

View Post

Google·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Google SWE coding round, graph traversal problem with BFS. Pretty standard but the string-keyed nodes tripped me up a bit on implementation.
  • Given a list of undirected relationships between named entities and a set of queries, determine for each query (s, t) whether s can reach t by following relationship edges.

“I knew it was BFS or union-find right away, but the string identifiers instead of integer node IDs slowed me down.”

View Post

Google·Software Engineer·Technical Phone Screen

Jun 2026
Google SWE coding round with a graph problem that looks straightforward until you realize there's a twist on what 'shortest total path' actually means when two nodes are both trying to reach the same destination.
  • Given an undirected graph with two nodes A and B, both need to reach node D. They can travel independently or meet at some intermediate node and continue together. Find the minimum combined path length for both to reach D.

“The example they gave was small enough that I almost just hardcoded a BFS and called it a day.”

View Post

Google·Software Engineer·Onsite - Multi Round

Jun 2026
Went through the full Google loop for a software engineer role, two virtual rounds then two onsite. Cleared the virtual rounds fine, but got rejected after onsite with a rejection reason that honestly made no sense to me. The recruiter cited 'no time for in-depth discussion' which felt completely backwards given how my rounds actually went.
  • A graph-based problem where the problem statement itself didn't obviously signal that graphs were involved. Solve it optimally.
  • Behavioral / culture fit round focused on values and ways of working.
  • Onsite coding question where the interviewer initially pasted an incorrect problem statement. After clarifying, solve the problem and then a follow-up variant of it.
  • A graph traversal or graph-related coding problem, with a follow-up extension requiring a second implementation.

“The disguise was the whole challenge here.”

View Post

Google·Software Engineer·Hiring Manager Screen

Jun 2026
Interviewed at Google for what seemed like a site reliability or engineering role. One question but it was the kind that makes you realize how much of this job is judgment under pressure, not just technical skill.
  • You're the only engineer available on a Friday afternoon and a request comes in to develop and deploy a production change. How do you handle it?

“My first instinct was to talk through the technical steps and I think that's where I lost the plot a bit.”

View Post

Google·Data Scientist·Technical Phone Screen

Jun 2026
Google DS interview, coding round focused entirely on a payment-to-invoice matching problem. Two parts, progressively harder. Felt more like a backend engineering screen than anything data science related, which threw me off a bit.
  • Given a list of invoices and a list of payments, write a function that parses each payment's memo field to extract an invoice ID (formatted as 'Paying off: <INVOICE_ID>'), looks it up in the invoice list, and returns the matched result or a clear not-found result. Handle edge cases like extra whitespace, empty memos, and malformed strings.
  • Extend the matching logic so that if the memo doesn't contain the standardized prefix, you fall back to matching by payment amount. If multiple invoices share the same amount, pick the one with the earliest due date. If none match, return an appropriate error message. Also state and justify the time and space complexity of your solution.

“I went straight to regex and the interviewer seemed fine with that.”

View Post