I sent my CV through the open roles portal for a Software Engineer (Backend) role.

A Sourcer reached out to schedule a 30-minute recruiter call. The call was to describe the role and give me the opportunity to discuss my background (ML PhD, somewhat unrelated to the role). After that, we scheduled a 1-hour technical screen (algorithms/data structures).

I got notified the same day that feedback was very positive and that they'd like to proceed to an onsite loop.

The onsite loop was scheduled about a week out, with two 1-hour interviews per day on Monday and Tuesday.

• Monday: Behavioral with Director of Engineering + Algorithms/Data Structures
• Tuesday: Coding + Systems Programming

I got notified on Wednesday that feedback was good again and that they'd like to proceed with reference checks.

I submitted references and synced with them about the process by Monday of the next week. The reference calls happened on Wednesday and Thursday.

Due to how the hiring committee works, it took until Tuesday of the next week before I got the confirmation they would be extending an offer. After that, it took a few more days to get a concrete offer, which I negotiated for about another week before signing.

The CoderPad platform was used for writing code, and Google Meet for the calls. Recruiters have access to a fairly complete list of recommended resources for the interviews, so make sure to ask if they don't send it over automatically.

The technical screen and algorithms round of the onsite were reasonably traditional.

Without giving any specifics (signed NDA), expect questions to be more "applied" than traditional LeetCode. For example, "We're trying to solve this concrete problem under the following constraints; how would you go about that?"

In my case, these questions involved a lot of follow-ups, to the point where I don't think you can "finish" the question in the allotted time. The interviewers kept making things more complicated until I got stuck or time ran out. In many cases, interviewers expect you to go beyond the best purely algorithmic solution and instead discuss tradeoffs, such as in my case, giving up resolution/granularity for improved speed.

The question during the onsite was significantly more involved than the one during the technical screen.

The behavioral interview was also fairly standard.

This was a one-hour chat with the Hiring Director of Engineering about my background, asking typical behavioral questions. After each answer, the interviewer would either ask follow-ups or recap my answer in a few sentences before moving on to a different behavioral question.

The coding round of the onsite loop is similar to an algorithms question, but the focus is not necessarily on providing a solution with optimal runtime. Instead, the focus is on making sure you can clearly think through a solution and write understandable code.

The systems programming round involved a question combining elements of computer science fundamentals, multithreading, and a big systems design component.