I kind of resent these attempts at optimally cherry-picking "the right" candidates using data.
Hiring is _intrinsically_ a subjective act.
People are very much a moving target. They change over time. Work experiences, even bad ones, shape one's skills and ability to cope in organizations. Almost everybody has a bad-fit job at one time or another. The experience of a bad-fit is actually important to the growth of the individual and, I think, their coworkers and employers.
This is the core way we look at it. Our models are hopelessly simplistic, and have no chance of modeling the true complexity of human ability. It's important to stay humble and recognize this. However, what we're competing against is people (often non-technical people) making gut screening decisions. And it ends up that even relatively simple models can do a better job than most recruiters making gut calls.
And the fact that we are able to do this while being blind to background (and open to people from all sort of backgrounds) is something that I think is very positive. We totally make mistake and reject good people. But we get better over time, and we also help lots of people get jobs they might not have been able to get without us.
There is a lot of scientific support for pre-employment screening (selection). The most conclusive evidence comes from a series of absolutely enormous studies conducted by the US military in the 80's known as project A.
That being said, there is a great deal of pseudo science being gobbled up by organizations because this is a mostly unregulated industry.
IIRC, the outcome of these studies were that intelligence is the best predictor of job performance. The ASVAB is essentially an IQ test with some domain knowledge questions sprinkled in.
When an organization scales to the point where they need something like TripleByte, they're going to be using some kind of impersonal screening process, though. This is better than some disinterested HR drone scanning resumes for keywords.
I agree that an HR drone scanning for keywords is not good (though some are better than others).
The most savvy candidates, however, tend to skip the step where you throw your resume into a giant vat with the others and hope for the best. Isn't it still the case that most jobs are filled using references and professional networks?
No, not at truly large companies -- Amazon could never fill its halls with just references and such.
Even at smaller companies, however, there's some pushback to network hiring because it isn't particularly great for getting a diverse pool of talent.
To be honest, though there are a lot of dangers with this sort of thing, I am a fan. End of the day, engineers should be great engineers, not great at hacking the job finding process.
there's some pushback to network hiring because it isn't particularly great for getting a diverse pool of talent.
I'd be surprised if random resumes are really that much more diverse than network hiring. Consider principles like the Erdos number[1] -- the collaborative distance between groups of people in similar fields is strikingly small. I bet, a company with a reasonably large development staff (50 people) is probably no more than 2-degrees separated from 99% of the talent pool in a given region.
Just about every position I've looked at was with a company where a former colleague works.
Comments
I kind of resent these attempts at optimally cherry-picking "the right" candidates using data.
Hiring is _intrinsically_ a subjective act.
People are very much a moving target. They change over time. Work experiences, even bad ones, shape one's skills and ability to cope in organizations. Almost everybody has a bad-fit job at one time or another. The experience of a bad-fit is actually important to the growth of the individual and, I think, their coworkers and employers.
This is the core way we look at it. Our models are hopelessly simplistic, and have no chance of modeling the true complexity of human ability. It's important to stay humble and recognize this. However, what we're competing against is people (often non-technical people) making gut screening decisions. And it ends up that even relatively simple models can do a better job than most recruiters making gut calls.
And the fact that we are able to do this while being blind to background (and open to people from all sort of backgrounds) is something that I think is very positive. We totally make mistake and reject good people. But we get better over time, and we also help lots of people get jobs they might not have been able to get without us.
There is a lot of scientific support for pre-employment screening (selection). The most conclusive evidence comes from a series of absolutely enormous studies conducted by the US military in the 80's known as project A.
That being said, there is a great deal of pseudo science being gobbled up by organizations because this is a mostly unregulated industry.
IIRC, the outcome of these studies were that intelligence is the best predictor of job performance. The ASVAB is essentially an IQ test with some domain knowledge questions sprinkled in.
When an organization scales to the point where they need something like TripleByte, they're going to be using some kind of impersonal screening process, though. This is better than some disinterested HR drone scanning resumes for keywords.
What point is that exactly? Google, FB, Apple, Amazon etc all rely on their own internal recruiting departments.
I agree that an HR drone scanning for keywords is not good (though some are better than others).
The most savvy candidates, however, tend to skip the step where you throw your resume into a giant vat with the others and hope for the best. Isn't it still the case that most jobs are filled using references and professional networks?
No, not at truly large companies -- Amazon could never fill its halls with just references and such.
Even at smaller companies, however, there's some pushback to network hiring because it isn't particularly great for getting a diverse pool of talent.
To be honest, though there are a lot of dangers with this sort of thing, I am a fan. End of the day, engineers should be great engineers, not great at hacking the job finding process.
I'd be surprised if random resumes are really that much more diverse than network hiring. Consider principles like the Erdos number[1] -- the collaborative distance between groups of people in similar fields is strikingly small. I bet, a company with a reasonably large development staff (50 people) is probably no more than 2-degrees separated from 99% of the talent pool in a given region.
Just about every position I've looked at was with a company where a former colleague works.
[1]https://en.wikipedia.org/wiki/Erd%C5%91s_number
But now you can reify your prejudices in a prior!