3 ms·
My understanding of Google is that they're ok with false negatives because they consider false positives to be far far worse.
by rongenre 7y ago
My understanding of Google is that they're ok with false negatives because they consider false positives to be far far worse.
- WrtCdEvrydy 7y agoFalse negatives are a burden on your candidates, false positives are a burden on your organization. You'd be surprised how often the answer to 'Should we hire' has more to do with 'He felt like he would be a dick' rather than 'He doesn't know X and Y technology'.
- 9nGQluzmnq3M 7y agoThat's the way it should be. A good candidate can learn X and Y, but it's hard to unlearn being a dick.
- babesh 7y agoI am not a Google employee, but from working with ex and subsequent Googlers and from interviewing Googlers, Google has some blind spots (perhaps intentional). It places more weight on algorithms (which can be gamed with study) and less weight on architecture, design, and coding. This sometimes results in candidates who can’t code well or engineers who badly overengineer. That said, some of the best candidates have been from Google. Perhaps getting those best people was worth a certain percentage of what I consider borderline false positives. Also, their ahole filter is imperfect. Just as with algorithms, you have people gaming the not an ahole filter.