The interview process consists of 5 rounds: 1 behavioral and 4 technical. This will vary depending on the position and level you're applying for.
The behavioral round is very standard. Look at Amazon's leadership principles for guidance.
The technical interview questions were LeetCode-style, although none were directly from LeetCode. They were easily the hardest questions I've ever encountered in any tech interview, including from Facebook, Amazon, Microsoft, Uber, and others.
Memorizing LeetCode won't be enough (as it would be for Facebook, for example). You'll need to really understand the underlying concepts and be able to apply them to do well.
After passing the technical interview, the recruiter will either:
Either way, you will need to both match with a team and pass the hiring committee to receive an official offer. You will also need to receive VP approval, although almost everyone who reaches this step will get approval.
Note that reaching the hiring committee does not mean you have an offer yet. In fact, from what I've heard, over 60% of candidates who make it to the committee are rejected by the committee.
But if you don't make it, you can always try again next time!
Problems related to graphs, binary search, and other algorithm/data structure concepts. I had never seen any of them before, although a couple were genuinely fun to solve.
The following metrics were computed from 96 interview experiences for the Google Software Engineer role in Seattle, Washington.
Google's interview process for their Software Engineer roles in Seattle, Washington is very selective, failing most engineers who go through it.
Candidates reported having very good feelings for Google's Software Engineer interview process in Seattle, Washington.