Excellent point; thought-provoking and well-put. Agreed completely.
The article does have a few good questions, and the best, IMO, is #36:
> 36. How would you build a product for people who are looking for an apartment?
That is such a good question because it's incredibly open-ended. There is no wrong answer. It doesn't even necessarily require a computer-based solution.
If the interviewer moderates the interview well, a single open-ended question with a broad scope can lead to wide-ranging conversations that shed light on more than just technical prowess.
Open-ended questions also present the interviewer the opportunity to ask spontaneous questions, digging into the thought processes behind a candidate's choices. That digging can and should yield lots of valuable, relevant information about the candidate. Assuming, of course, the interviewer has the skills necessary to extract that value from the interview.
Ask "Why?" enough in these situations, and you eventually get something close to the truth[1].
Yes, this is my favorite kind of interview question. These days my goal as an interviewer is more to find somebody's strengths than their weaknesses. Nobody knows everything; nobody's good at everything. So what I'm trying to figure out in an interview is what this person can add to the team.
Sure, you do need to keep an eye out for red flags. But you won't find all of those in an interview anyhow, because people can hide those. It's much harder to fake a strength, and it's their strengths are ultimately what we hire people for.
Way more engineers fail these, IME, than algorithms questions. I have to have a pretty low bar for "willingness to speculate and have open-ended conversation about approaches to problems," unfortunately.
Could you define what you mean by "ace" those questions?
I don't think giving a smooth or glib answer would be acing it at all, for instance.
But someone who gets super defensive, or pissed off, that easily by questions in an interview probably isn't someone who you'd want in a managerial or high-degree-of-communication role (like even a sufficiently senior IC role). Talking to people, and convincing people, even people who don't like you, is part of the job, there.
Of course, if you're the candidate who doesn't like that stuff, and the interviewer is trying to dig on that, it's probably a good sign that the skills they expect you to be using in the role aren't the ones you want to be using.
- I was at the company for a decade and it was time to move on to bigger challenges.
- I didn’t leave the job, the job left me - they went out of business - but I survived multiple rounds of layoffs and even afterwards I went to work as a contractor for one of our clients
- the product they initially hired us for wasn’t successful and we were being moved to maintained a legacy PHP product.
- I was bought in to bring an outside perspective to the company but we were in a satellite office and our influence was limited.
- the company changed direction and decided not to be a software company.
ask the question, but simply choose the answers with the rough edges that might merit a bit more thought and discussion vs those that 'ace' the answers.
It's a lousy hiring indicator.
Having been in panels where we had to ask some of those specific questions (specifically for hiring managers), and then seeing how those managers acted later on the job, the very worst hires (manipulative people who have no interest in the company as a whole, but gaining power) always ace those questions. The best hires of the lot, in practice, tended to give worse answers with some red flags, and only double checking on said red flags showed that yes, the rough edges in the answers were absolutely justified by their experiences.
So I'd not ask those questions myself when I can help it, as I'd rather not select for the best available manipulator.