Sometimes, the vocal majority on hn is highly skeptical, going above and beyond the call of "duty" to point out unsupported conclusions, anecdotal evidence, etc. Other times, they grab their pitchforks and attack the enemy (government, management, corporations, patent law) without so much as a glance at the data.
1. Are you certain that all technically qualified candidates will fit in with every company? Do you have data that supports this opinion?
2. Assuming there are some "poor fits", are you certain that the HR department and its tests are unable to accurately detect poor fits? Do you know what the sensitivity and specificity of the tests are?
3. Given that there will be some false positives (incorrectly labeled as "poor fit"), are you certain that losing these false positives harms the company more than hiring the true positives would (because you don't detect them)? Do you have data that supports this opinion, for various objective measures of harm or success?
Consider that these programs may be data-driven, which your anecdote is not, and therefore a) this person might actually be a technically qualified yet poor fit for the company and you just can't tell and b) this person could be an acceptable false positive, and that judging a program due to a specific instance is vacuous. My gut reaction is that, probably, the best hiring program allows several people and departments with various expertise should each have veto power.
Finally, hackers are traditionally anti-authoritarian, and that is both good and bad. Failure to realize that, and to admit that authority may have some value that even you, brilliant computer programmer, can't personally vouch for, is one of the aspects of our community that I would love to see change.
1) Absolutely not, there's no perfect way to figure out if someone is a good fit for the company.
2) I expect that the HR process is evidence towards whether a candidate is a good fit, but not as good evidence evidence as the estimation of future coworkers regarding if they would be a good fit. I expect that written tests by HR regarding personality would be approximately useless, and that the people who created them haven't measured their sensitivity and specificity.
3) I expect that would vary wildly case by case.
EDIT: To unpack 2 a bit more, I'm sure there are personality tests that would be fairly useful in these cases, but I also expect that actual researchers who are doing things scientifically and only promise what they can deliver would be at such a huge disadvantage selling to HR departments that I wouldn't expect any uptake even without competition from unscrupulous types who think they've got everything figured out and don't feel the need to scientifically test their methods.
1. Are you certain that all technically qualified candidates will fit in with every company? Do you have data that supports this opinion?
2. Assuming there are some "poor fits", are you certain that the HR department and its tests are unable to accurately detect poor fits? Do you know what the sensitivity and specificity of the tests are?
3. Given that there will be some false positives (incorrectly labeled as "poor fit"), are you certain that losing these false positives harms the company more than hiring the true positives would (because you don't detect them)? Do you have data that supports this opinion, for various objective measures of harm or success?
Consider that these programs may be data-driven, which your anecdote is not, and therefore a) this person might actually be a technically qualified yet poor fit for the company and you just can't tell and b) this person could be an acceptable false positive, and that judging a program due to a specific instance is vacuous. My gut reaction is that, probably, the best hiring program allows several people and departments with various expertise should each have veto power.
Finally, hackers are traditionally anti-authoritarian, and that is both good and bad. Failure to realize that, and to admit that authority may have some value that even you, brilliant computer programmer, can't personally vouch for, is one of the aspects of our community that I would love to see change.