3 ms·

Here's the thing though. It's a lot more effort to look through projects on GitHub and evaluate them. And even if you can, you then have to figure out how to cr

Here's the thing though. It's a lot more effort to look through projects on GitHub and evaluate them. And even if you can, you then have to figure out how to cross compare that and set a consistent bar vs people who you interview because they don't have a GitHub presence. People want hiring processes to be somewhat consistent, and in that world it's really kind of a non-starter to replace tech screens with GitHub profile reviews.

tessierashpool 6y ago
yup. and GitHub's useless for value adds like management, marketing, recruiting, or product skills. for that matter, GitHub is useless if the projects you contribute to are on BitBucket or GitLab. that contributions graph has a lot of assumptions baked in.
hinkley 6y ago
There is not that much code I wrote three, four years ago that I’m still proud of. I figured that would get better with age but it’s only by degrees, and at any rate it may indicate that I’m getting soft rather than better. So anything on github that I might want you to look at? I’d have to be on a three year replacement cycle, permanently, which is a hell of a lot more work than you might estimate that a reasonable github presence would entail.
asdff 6y ago
It's so backward to see companies shift to these abstracted whiteboard or homework assignments as some sort of standard, considering standardized testing is now falling out of favor in higher education as it is rarely indicative of actual potential success. It strikes me as lazy, just following a trend unrooted in any empirical data. Maybe the solution is for companies to hire more thorough managers who are capable of vetting a github profile to see if the candidates skills are relevant to the position, not turn the interview process into a performance.
roland35 6y ago
I think it is also hard to compare different candidates this way, even assuming they both have a similar level of commitment on GitHub. A take home exercise at least allows them to compare the same thing

tessierashpool 6y ago

yup. and GitHub's useless for value adds like management, marketing, recruiting, or product skills. for that matter, GitHub is useless if the projects you contribute to are on BitBucket or GitLab. that contributions graph has a lot of assumptions baked in.

hinkley 6y ago

There is not that much code I wrote three, four years ago that I’m still proud of. I figured that would get better with age but it’s only by degrees, and at any rate it may indicate that I’m getting soft rather than better. So anything on github that I might want you to look at? I’d have to be on a three year replacement cycle, permanently, which is a hell of a lot more work than you might estimate that a reasonable github presence would entail.

asdff 6y ago

It's so backward to see companies shift to these abstracted whiteboard or homework assignments as some sort of standard, considering standardized testing is now falling out of favor in higher education as it is rarely indicative of actual potential success. It strikes me as lazy, just following a trend unrooted in any empirical data. Maybe the solution is for companies to hire more thorough managers who are capable of vetting a github profile to see if the candidates skills are relevant to the position, not turn the interview process into a performance.

roland35 6y ago

I think it is also hard to compare different candidates this way, even assuming they both have a similar level of commitment on GitHub. A take home exercise at least allows them to compare the same thing