4 ms·

Vital meta lesson from this: I was rejected by Fashionable BigCo X and Y but then eventually got a job at Fashionable BigCo Z When we see that repeated for ma

Vital meta lesson from this: I was rejected by Fashionable BigCo X and Y but then eventually got a job at Fashionable BigCo Z When we see that repeated for many successful people, I hope the takeaway becomes: your interview process is probably not doing what you think it is. In some cases this extends to the entire work culture. E.g. workplaces that situate themselves in a mindset equivalent to "we must hire the best unicorns to possibly succeed". Other, better ones instead try to embody "we will continually improve our hires. we will help make them great. (oh, there are no unicorns.)"

bradleyjg 8y ago
Do you suggest always hiring the first person that applies? There has to be some choosing function.
- saidajigumi 8y ago
  No. But I think many of our filters are bogus. Some of these are easy: "Did you complete a traditional four-year CS degree at <highly respected university>?" should be considered downright obscene as a front-line filtering function. It robs your team of great talent, and destroys any attempt to hire diverse candidates. And by "diverse" I mean invaluable team members who will bring fresh perspective to your problems, your understanding of your users, and more. There have always been many "alternative" paths into computing, now moreso than ever. A concrete example: A place I interviewed in my last search uses pair programming extensively in their day-to-day (and have lovely twinned monitor/keyboard hardware setups to support this!) They use that in their tech interviews as well. You and a candidate sit and talk about code. It's a lovely experience, and far more grounded in the work than just about any whiteboard interview. Pair-programming isn't the takeaway here. It's that this org built an interview model which was virtually identical to their actual working model.
  - bradleyjg 8y ago
    I agree 100%. I myself don’t have a CS degree. I thought though with your emphasis on training great employees rather than finding them you were talking about hiring people that maybe aren’t the great programmers. It’s hard enough to design a selection process for exactly what you are looking for (i.e. programming ability), I have no idea how one would go about identifying candidates with great potential that can’t already program.
cataphract 8y ago
Any hiring process will result in some good employees being rejected and bad employees being accepted. If your candidate pool is very large compared to the open positions, as is the case for the Fashionable BigCos, it makes sense to "tighten the mesh" in order to avoid the false positives better.

bradleyjg 8y ago

Do you suggest always hiring the first person that applies? There has to be some choosing function.

saidajigumi 8y ago
No. But I think many of our filters are bogus. Some of these are easy: "Did you complete a traditional four-year CS degree at <highly respected university>?" should be considered downright obscene as a front-line filtering function. It robs your team of great talent, and destroys any attempt to hire diverse candidates. And by "diverse" I mean invaluable team members who will bring fresh perspective to your problems, your understanding of your users, and more. There have always been many "alternative" paths into computing, now moreso than ever. A concrete example: A place I interviewed in my last search uses pair programming extensively in their day-to-day (and have lovely twinned monitor/keyboard hardware setups to support this!) They use that in their tech interviews as well. You and a candidate sit and talk about code. It's a lovely experience, and far more grounded in the work than just about any whiteboard interview. Pair-programming isn't the takeaway here. It's that this org built an interview model which was virtually identical to their actual working model.
- bradleyjg 8y ago
  I agree 100%. I myself don’t have a CS degree. I thought though with your emphasis on training great employees rather than finding them you were talking about hiring people that maybe aren’t the great programmers. It’s hard enough to design a selection process for exactly what you are looking for (i.e. programming ability), I have no idea how one would go about identifying candidates with great potential that can’t already program.

cataphract 8y ago

Any hiring process will result in some good employees being rejected and bad employees being accepted. If your candidate pool is very large compared to the open positions, as is the case for the Fashionable BigCos, it makes sense to "tighten the mesh" in order to avoid the false positives better.