undefined | Better HN

0 pointstptacek13y ago0 comments

Yes. That's what I'm saying.

I'm not saying that (eg) Google has a better option than conventional interviews; Google has unique hiring problems. That doesn't make the interview system they use better. (Also, I don't have great intel on how Google interviews work; nothing better than you have. Maybe they've solved this problem in other ways... though, I'm skeptical.)

Startups, though, have no excuse.

0 comments

2 comments · 1 top-level

pfedor13y ago· 1 in thread

So do you believe that if you take a random sample of (eg) Google engineers, and a random sample of all software engineers, there will be no difference in ability between the two groups? Or do you think there would be a difference but attributable to something else than their hiring practices?

tptacekOP13y ago

I think conventional job interviews do allow you to target a broad range of competence, with a large margin of error, so if you were to compare Google employees to, say, State Farm line-of-business .NET application developers, sure, the Google engineers would outperform.

It's an interesting question whether a random sampling of Google engineers would outperform a random sampling of tech industry product developers from other companies. I don't know.

The bigger is, within the broad set of candidates that an interview process selects for, the conventional job interview is arbitrary bordering on capricious; it's inhumane for no business benefit.

j / k navigate · click thread line to collapse