← Back to Directory

Google interview notes, page 165 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·Hiring Manager Screen

Jun 2026
PM interview at Google with a policy-focused question about remote work. Pretty light on technical content, felt more like a culture/strategy screen than anything else.
  • What changes or improvements would you suggest to the company's work-from-home policies to boost employee productivity and satisfaction?

“I went straight into async communication tooling and flexible core hours, which felt like the obvious angle.”

View Post

Google·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Google system design round, one big question about migrating a relational DB to a NoSQL key-value store. Covered a lot of ground and felt like I was sprinting the whole time.
  • Design a system to migrate data from a relational database to a NoSQL key-value store, starting with a consistent snapshot and then switching to change data capture via the write-ahead log.
  • How do you guarantee snapshot consistency, and how do you handle the ordering boundary between the snapshot and the CDC stream?
  • How would you achieve exactly-once semantics across replays and handle failure recovery?
  • How do you handle schema evolution during and after the migration?
  • Walk me through how you'd partition the data and queue the migration work, and what monitoring you'd put in place.

“This was basically the whole interview.”

View Post

Google·Product Manager·Onsite - Product Sense / Strategy

Jun 2026
Google PM interview, product strategy question about market selection. One question, pretty open-ended, felt like they wanted to see how you structure ambiguity more than get a right answer.
  • You're launching Google Maps in Argentina vs. Brazil. How would you decide which market to enter first?

“I went straight to market size and blanked on the more interesting stuff for probably two minutes.”

View Post

Google·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Google SWE coding round with a board game simulation problem. The problem looked deceptively simple but the state space caught me off guard once I started thinking about multiple tokens interacting.
  • You have a 1D board represented as a string with empty cells, tokens, and coins. Each token can move exactly 3 positions to the right, cannot land on another token, and collects a coin if it lands on one. What is the maximum number of coins collectable?

“My first instinct was greedy and it was wrong.”

View Post

Google·Machine Learning Engineer·Onsite - System Design / Architecture

Jun 2026
Google ML engineer interview with a system design angle focused on improving an existing system using your area of specialization. Pretty open-ended, which sounds nice until you're actually sitting there trying to scope it.
  • Pick a system and propose meaningful improvements to it using your ML specialization, whether that's recommendations, NLP, or computer vision.

“The open-endedness is what gets you.”

View Post

Google·Software Engineer·Onsite - Coding / Algorithms

Jun 2026
Google SWE coding round with a tree problem that looked like a BST variant but had a twist I didn't fully appreciate until halfway through. The follow-up about broken trees was where things got interesting.
  • Given the root of a trinary search tree where each node has a left child (values less than current), a middle child (values equal to current), and a right child (values greater than current), find the mode or modes of the tree. Return all values tied for highest frequency.
  • How would your approach change if the tree structure is broken, meaning nodes may not actually respect the less-than, equal, greater-than ordering constraints?

“My first instinct was to just do a full traversal and count with a hashmap, which works, but I kept second-guessing myself thinking they wanted me to exploit the tree structure somehow.”

View Post

Google·Software Engineer·Technical Phone Screen

Jun 2026
Interviewed at Google for a machine learning or data science role, got hit with an ensemble methods question that felt more theoretical than I expected for a phone screen.
  • How does the way boosting methods combine trees differ from how random forests do it?

“I knew the surface answer but fumbled explaining the sequential vs parallel distinction clearly.”

View Post

Google·Data Scientist·Technical Phone Screen

Jun 2026
A technical screen for a Data Scientist role at Google that went deep fast. The whole thing was basically one extended ML design problem about URL classification, and they kept pushing on evaluation methodology and deployment strategy in ways I wasn't fully prepared for.
  • You have 1,000 labeled URLs (good/bad) and a large unlabeled pool where bad links are rare. How would you design features and train a logistic regression model for this?
  • With severe class imbalance in this dataset, how would you evaluate your model? Walk through your full evaluation plan.
  • How would you choose a decision threshold for this classifier given that misclassification costs are asymmetric?
  • What approaches would you use to handle class imbalance during training, and what are the tradeoffs?
  • How would you check for data leakage in this setup, and what would you monitor once the model is in production?
  • Describe how you'd validate this model offline before deploying it, and what a staged rollout would look like.

“I jumped straight into feature ideas (URL length, domain age, TLD type, presence of subdomains, etc.) and got through the modeling part okay.”

View Post

Google·Software Engineer·Technical Phone Screen

Jun 2026
Two coding problems in what felt like a pretty standard Google SWE round. Both had follow-ups that pushed you further, which I wasn't fully ready for.
  • You have an array representing link weights in a gold chain. Remove exactly one link, reconnect the rest in order, then make a single cut to split the chain into two non-empty parts with equal total weight. Return true if any valid (removed index, cut position) pair exists. Follow-up: return all such valid pairs.
  • Given an integer array, find indices i and j where the values at both endpoints are equal and the subarray sum from i to j is maximized. Return one such pair. Follow-up: can you do it in O(1) extra space?

“Prefix sums are the obvious tool here and I got there pretty fast.”

View Post

Google·Software Engineer·Technical Phone Screen

Jun 2026
Interviewed for a BA role at Google and got hit with a data integration question that felt more like an engineering problem than anything I'd prepped for.
  • How do you handle joining data from different sources when the IDs don't match up consistently?

“Wasn't expecting something this technical in a BA screen.”

View Post