[stlHusker]

I really hope that more people would look at it that way.

Just my own personal experience, but I've honestly seen the trial-by-fire method used either at the sole exclusion of, or weighted more heavily than actually talking to the candidate about their past work. I can't count the number of times in the past where I've encountered this.

At most points when this occurs and it seems there is no end in sight, I typically state: "You know, I've solved some pretty interesting problems over my career, why don't you ask me about that, I think problem xyz it is similar to what you are looking for; let me tell you why". Subsequent laughs and eye-rolls later you get a canned: "I don't care about that, I care about things that matter to me" response.

When you start selecting "problems" for candidates to work through, you add additional layers of interference into the activity: 1) Have you selected the right interviewer for this type of questioning line? Can they describe a the problem adequately? Can they ask probing questions? Are they only able to follow one line of thought? Are they patient? Are they arrogant? 2) Does the problem test for what you think it does? 3) Is the candidate able to think on their feet on a minutes notice on a topic they may not be familiar with in front of an audience?

Where I work (aerospace, embedded systems) interviews are ~75% behavioral and ~25% "technical" with a major focus on having the candidate work through problems they have solved in the past, with the interviewer asking questions not to break but to really learn about them.

I have family members that work at various levels within the medical profession, none of them get the "generic problem solving" trial-by-fire experience when interviewing. (Yes, I realize there is a higher level of trust there, certifications, boards and all. I just think it is a bit crazy that we cannot find a better way).

[extr]

This reminds me of an data science interview I had early in my career that was exactly along these lines. Several "case studies" that turned out to be semi-guided math/cs interview problems. One guy started off on a variation of the secretary problem, I recognized the general form of it and pointed out that there is a straightforward way to derive the optimal solution, which we could tweak for our situation. No no, we have an hour to kill. I want you to explain my way of solving the problem back to me. Cue an hour of frustrating whiteboarding where neither I nor the interviewer seemed to really grok the intuition for the steps we were taking. I went through 3 more situations like this over the course of a grueling 8 hour day, probably about an hour of which I had the chance to talk about my own background. Didn't get the job, and was specifically called out for my "poor technical performance". I had been through so few interviews at that point I had no idea it wasn't the norm. I legitimately wondered if I was cut out to be in the profession. Felt like major shit.

A little while after I had an interview more like what you describe. It was night and day. We did a brief technical portion that was mainly to ensure I wasn't bullshitting my experience level, then a few hours of discussing my past projects with various current team members. I actually enjoyed it, found myself relating to my interviewers, and easily got the job.