4 ms·

Fancy moving to the UK? :) Seriously there's a job with your name on it if you're interested. When I interview candidates I sit with them for half an hour or

Fancy moving to the UK? :) Seriously there's a job with your name on it if you're interested. When I interview candidates I sit with them for half an hour or so to get to know them. Then I give them purposely broken, poorly written piece of code which I tell them to pull apart. This proves incredibly effective as even if they miss some of the more obvious errors I can at least point them in that area and then see if they can see the problem on their own. There are about 100 different things to talk about so it really gives me an idea of the level they are at, and also the type of programmer they are; passionate, lazy, smart, meticulous, inexperienced, confident etc. Then if I feel they are worth a second interview, I get them back to sit with me and my team for the day to see how they fit in with the team. Then all being well I offer the job.

koyote 10y ago
That sounds pretty much like an ideal interview process. But do you think this is easier to do if you get less candidates (looking at your location, I am assuming you only get a small fraction of candidates than if you were located in, say, London) ?
- m4tthumphrey 10y ago
  Yes I do. We have actually found it very hard to find good people to interview let alone prvovide an offer. But nevertheless, I'm not sure why this wouldn't work when interviewing more candidates.
- shawn-furyan 10y ago
  > But do you think this is easier to do if you get less candidates This is one of my pet peeves. The number of applications you get has absolutely nothing to do with the number of candidates you should consider, because it has absolutely nothing to do with the number of candidates that you can competently consider. You don't have to get through every resume, and trying to do so will just lead to biased "how can I say no" ad hoc filtering games. The solution is to use an unbiased* pre-filter to restrict the number of applicants that you let into a thorough consideration process. Random selection is the pre filter that I would use, but there may be other valid choices (though outside of random selection, it's easy to let bias slip in). Bonus points if you are transparent with candidates that get filtered out this way about why and how that happened. If you're messaging is constructed well, then they should understand that there were too many candidates to give all of them fair consideration, and that being filtered out reflects in no way on their worth as a candidate. So, they will be more likely to apply for future openings. Sorry if it feels like I'm criticizing you. I don't mean to at all. It's just that this idea of being overwhelmed by candidates is absolutely pervasive in discussions of hiring, and it's a completely fake problem that leads to absolutely real difficulty on both sides of the hiring process. * A lot of organizations resort to biased pre-filters to cull resumes, like unnecessary requirements. I'm of the opinion that this leads to suboptimal results, particularly if they align significantly with the biased filters that the rest of the industry your company is competing in uses. It means that there are qualified candidates that end up systematically under-recruited. Those are the Moneyball candidates. You should want those candidates because they are the most productive relative to the salaries they can demand. And even if you don't give them a lower salary than you would give a candidate that passes all the typical filters, then you will still have a significant retention advantage, which all told may be worth more than shaving salary in many (most?) situations.
1stranger 10y ago
You talk to them, like they're people or something? What is this witchcraft? But seriously, I do the same thing. Have them look over some broken code for 10 minutes to collect their thoughts then have a discussion about the issues they found and various tangent topics relating to it. I'm amazed how uncommon this is.

koyote 10y ago

That sounds pretty much like an ideal interview process. But do you think this is easier to do if you get less candidates (looking at your location, I am assuming you only get a small fraction of candidates than if you were located in, say, London) ?

m4tthumphrey 10y ago
Yes I do. We have actually found it very hard to find good people to interview let alone prvovide an offer. But nevertheless, I'm not sure why this wouldn't work when interviewing more candidates.
shawn-furyan 10y ago
> But do you think this is easier to do if you get less candidates This is one of my pet peeves. The number of applications you get has absolutely nothing to do with the number of candidates you should consider, because it has absolutely nothing to do with the number of candidates that you can competently consider. You don't have to get through every resume, and trying to do so will just lead to biased "how can I say no" ad hoc filtering games. The solution is to use an unbiased* pre-filter to restrict the number of applicants that you let into a thorough consideration process. Random selection is the pre filter that I would use, but there may be other valid choices (though outside of random selection, it's easy to let bias slip in). Bonus points if you are transparent with candidates that get filtered out this way about why and how that happened. If you're messaging is constructed well, then they should understand that there were too many candidates to give all of them fair consideration, and that being filtered out reflects in no way on their worth as a candidate. So, they will be more likely to apply for future openings. Sorry if it feels like I'm criticizing you. I don't mean to at all. It's just that this idea of being overwhelmed by candidates is absolutely pervasive in discussions of hiring, and it's a completely fake problem that leads to absolutely real difficulty on both sides of the hiring process. * A lot of organizations resort to biased pre-filters to cull resumes, like unnecessary requirements. I'm of the opinion that this leads to suboptimal results, particularly if they align significantly with the biased filters that the rest of the industry your company is competing in uses. It means that there are qualified candidates that end up systematically under-recruited. Those are the Moneyball candidates. You should want those candidates because they are the most productive relative to the salaries they can demand. And even if you don't give them a lower salary than you would give a candidate that passes all the typical filters, then you will still have a significant retention advantage, which all told may be worth more than shaving salary in many (most?) situations.

1stranger 10y ago

You talk to them, like they're people or something? What is this witchcraft? But seriously, I do the same thing. Have them look over some broken code for 10 minutes to collect their thoughts then have a discussion about the issues they found and various tangent topics relating to it. I'm amazed how uncommon this is.