Recently interviewed for HubSpot in Dublin for a Senior Software Engineer position. The process was straightforward.

It started with a screening call with a recruiter, who was very good. This was followed by a take-home technical assessment involving pulling and posting data to an API. I managed to complete this quite quickly (around 30 minutes, I think), but candidates are given 3 hours if needed.

Then came the onsite interview, which consisted of two system design interviews and one coding interview. The interviewers were pleasant, and the questions were not overly difficult. I was able to finish the system design and coding questions in a timely manner, providing good explanations and justifications for each step. There were no noteworthy hiccups or even second thoughts on anything. In my opinion, the interviews had gone exceptionally well, much better than I would have expected to pass.

I was quite surprised to receive a rejection. I was able to request feedback, and another call with the recruiter was scheduled, which is great as I'm sure many companies don't offer specific feedback.

It kind of goes downhill from here. The specifics of the feedback were a bit frustrating and, to be honest, felt like a red flag. The feedback mostly consisted of extremely benign points that you'd never even consider mentioning to a candidate. One of the main pieces of feedback from the coding interview was not using getOrDefault for a Java HashMap, instead opting to do it in two lines (I had actually originally pointed out that I preferred the two-line solution to the interviewer and mentioned that I know it can be done in one line). I was very perplexed to see these were the points being used as make-or-break feedback.

Some other feedback points included spending too long explaining the time and space complexity of the solution rather than stating it immediately (which is sort of the opposite of how you are typically expected to relay this information in interviews). There were other minor points of feedback, but they all felt like moot points, or at least things you'd purposely do in an interview setting.

These sorts of points are quite orthogonal to the industry-standard rubric for judging tech interviews, where you would be expected to over-explain and take your time outlining your thoughts, and would typically not be nitpicking on code syntactical sugar choices. Sadly, despite receiving specific feedback, there don't seem to be any actionable points. In the sense that, if I did the interviews again, I'd do them the exact same even knowing the questions in advance because the feedback didn't really have anything to "improve" on in that regard.

I felt a bit bad for the recruiter because she was obviously just relaying the feedback and likely doesn't understand how preposterous that feedback actually is to give to a candidate. It's borderline like telling the candidate you didn't pass because they didn't like your haircut.

So, all in all, it was a good recruiter and a good interview experience, but the rubric on which the interviews seem to be judged appears to be a bit of a broken process. I've taken a quick look on Glassdoor post-interview and have actually noticed similar cases, and it seems to be accurate. I don't get the impression that the specific interviewers are to blame, but more so the style of rubric they are told they must judge the interview by. It was a bit of a letdown in the end, considering how well the interviews went.

2 system design questions and 1 coding question.