4 ms·
Sure. - What's your hit rate? (Detected an actual bullshitter) All of them, I definitely had a few bad hires, but never due to underestimating capability, als
by endymi0n 7y ago
Sure.
- What's your hit rate? (Detected an actual bullshitter)
All of them, I definitely had a few bad hires, but never due to underestimating capability, also most of them were from the early days.
- What's your false alarm rate? (Labelled someone a bullshitter who wasn't)
I see no way to come up with any good number for that, but I think at ~10% hiring rate from the interview I'd call myself pretty average from a perspective of selectiveness at that part in the funnel.
> This is again pretty easy to BS.
This is not the point of the question, I could care less whether 10% are slightly beautified. My focus with this questions and the follow ups are whether candidates can reproduce insights, learnings, hacks and fulfillment of business objectives in a logical and coherent way. If the story is fabricated, it usually falls apart when I ask "why" and "how did you actually do it?"
It's hard to fake expert knowledge about the stability of current sources when I drill down on the details of a candidates' recent Arduino project.
- omginternets 7y agoIt's interesting and revealing that you (deliberately?) fail to mention how you arrived at these numbers. It's almost as if you have absolutely no clue how good you actually are at catching bullshitters.
- perl4ever 7y agoThat's not fair or accurate. They said they had a basis for concluding they didn't fail to catch BSers - experience with hires afterwards. They also said they could only guess at the number of false positives. That means they supported their conclusion where they made one, and refrained from a definite conclusion where they couldn't make one.
- omginternets 7y agoYes, and that basis is based on ... nothing! Actually, it's worse. The claimed 100% hit rate implies classifying everybody as a bullshitter.