Given the usual nature of coding screens - intentionally generic and standardized, so a company can use the same screening process for different teams - it made a lot of sense to me that a task like that could be outsourced. Since you are usually not developing a relationship with the interviewer at that point, there’s no real reason they need to be and employee of the company you are interviewing for. A couple of times I’ve had coding screens where the interviewer didn’t seem to be paying attention. Since some engineers get forced to do these screens, they might not be focused on giving a good candidate experience.
It actually makes a lot of sense to do it this way on both ends.
You know the interviewer is going to be a good interviewer who knows what they are doing & how to judge you. The interviewer can be completely independent from the decision without having bias or judgement attached to where the candidate ends up.
I've certainly had interviews where I knew I was doomed from the second I walked in the door because the interviewer was in a bad mood, they didn't enunciate clearly, just a bad vibe, whatever. That sucks as a candidate. This situation would make a fairly demeaning process slightly better if done well.
The lucrative cottage industry of interview assessments won't be going down so soon as long as there are perverse incentives and gatekeeping in place.
Even people who got declined usually still left feeling quite positive, because we would turn the obviously-failed interview into, basically, a free hour coaching / teaching session for them to succeed next time.
Further - you guys lurk all over HN commenting like this every time there's an interview thread. It's submarining and it's gross.
One thing I didn't appreciate with these rules from a candidate perspective (and encouraged my decision not to continue my onboarding with Karat), there seems to be no way to learn from the process. I actually bombed my practice interview (and knew it would happen even before the interview) but the interviewer was very reluctant (due to the above rules) to actually solve the problem with me, where as a conventional non-Karat failed interview would at least give me the solution in the end and I'd at least learn something.
Still I don't know if I would trust hiring to unknown potentially awful interviewers.