Comment by golly_ned
1 day ago
Speaking from my experiences in dozens of debriefs: I don't think I can recall a case where an otherwise good candidate was rejected for something like not liking someone's personality.
There might be a middle ground between "technical skills" and "personality", though, which we do take into account, falling under "soft skills", which may be affected by certain personality traits. Things like polling the interviewer for their thoughts, asking thoughtful questions, being curious about the source of disagreement or misunderstanding, not being dogmatic, and so on. I think it can be harder to demonstrate these kinds of skills with certain personality types -- I used to be very nervous in interviews, and it wasn't always easy to have the presence of mind to exercise these soft skills.
But even still, at least in technical interviews like programming or system design (as opposed to cross-functional/manager/tech leadership interviews), I've found it relatively rare for a candidate to be rejected for 'soft skill' failures when the right signals are there for technical strength.
I just think it has to be soft skills in a few particular cases; I always check my work after the interview to see if I got the questions right, almost always by writing a PlusCal spec and checking against the invariants that they specified. Sometimes my answer is wrong and doesn't satisfy the invariant and there's edge cases that I didn't consider, and when that's the case I'm not really surprised when I'm rejected.
But when I get the technical questions correct and TLA+ seems to agree with me, that's where I get really confused. I'm not typically interviewing for management positions, I'm interviewing for engineering jobs.
I would counter, in that depending on the position(s) and candidate pool soft skills will make all the difference in the world when choosing between two technically sufficient candidates. It may not even be a conscious bias, so much as, "I'd rather work with this person."
Most technical jobs can be done by most sufficiently skilled, or motivated candidates with enough intellect to handle the work. Not always to a certain level of craft, efficiency or trend setting, but it's not about that to most business stakeholders. I say this as someone who deeply cares about the craft, despite a rather insulting depiction, including some of my opinions and age I came across earlier today (grugbrain.dev).
Likeability is hard, and even then striking a balance in a given context is also hard. It often comes down to a level of self-reflection, which is where I think TFA is at right now. Which is an attempt to establish a balance of personal responsibility, with "culture fit." Given that many of us have personality traits that tend to be deviations from the norm in many ways, it's all that more important to understand that in ourselves and our efforts to adapt to society.
A pinnacle of this in Television is the show Dexter, where he regularly brings in doughnuts. I've had coworkers that did similar, and it's impressive the amount of affect this has on the working environment and relationships in turn.
Beyond this, comes the counter-intuitive position of being far more skilled than an interviewer. This can definitely work against you at times as well for a number of reasons. Just because you are the most technically adept for a position, doesn't mean you get the position. The technical aspects of most jobs are more about a minimum of, can they do the job. Not, are they the best fit.
> Beyond this, comes the counter-intuitive position of being far more skilled than an interviewer. This can definitely work against you at times as well for a number of reasons. Just because you are the most technically adept for a position, doesn't mean you get the position. The technical aspects of most jobs are more about a minimum of, can they do the job. Not, are they the best fit.
This has been happening to me a bit, at an increasing frequency lately
I don't want to come off as cocky, but I think I have a decent understanding of concurrency and distributed systems, and I think a lot of interviewers simply do not. Sometimes I'm "corrected" on my whiteboard examples, and I have to push back because I'm not actually wrong about something.
And that kind of makes me seem like a douchebag, but at the same time I'm not going to pretend that I'm wrong on something if I don't genuinely think I might be wrong.