I have a very small consulting company. Both of my employees right now were non-traditional students when they finished their CS degrees. Actually, I'm not even sure the one guy has finished his degree yet. Regardless, neither of them had their degrees when they started working for me.
And honestly, I didn't really even know. I never asked them for a resume and they never provided one.
I don't really test the people I hire. I get to know them on a personal level. I meet them at parties, at meetups, conferences, etc. I then ask if they'd like to make a little side money with a freelance gig. We talk about the work, and if the person says they can do it, I give them a task. A real task, it goes right into the project I'm working on. It usually takes about a week or two.
And if they do a good job, I ask them to do more. If they don't, I pay them for the work they did, thank them, and don't mention they were ever interviewing for a more long-term relationship.
This has been so incredibly much better of an experience than the hiring processes I was involved with as a working stiff. Resumes that don't say shit. Stupidly simplistic coding tests. Interviewing anxiety. All of it, a nightmare.
And it didn't seem to make any difference. We would put more and more effort into hiring, to the point I was spending 10 to 15 hours interviewing and reviewing candidates a week, for weeks on end. It was having a detrimental impact on my performance, without any correlated impact on the performance of the people we hired.
So when I went independent, I decided that, if the results are really, actually random, then that means the inputs have no impact. So I threw away the interviewing process. I just ask people, "can you do this job"?
And really, I don't need to know anything else. That's what it all boils down to. They can lie on their resume just as much as they can lie to the question. They can cram on interview questions the night before. They can be really good at regurgitating academic theory but complete crap at thinking creatively. So I just ask, "can you do this job?"
I think this method doesn't work in all situations. For example in embedded programming. First there are not many of those who actually work in this field so unlikely to just meet up with them. Most are stabily employed. Secondly, there is lots of proprietary hardware and knowledge. Third, hard to work on a trial job when you are already working. Lastly there are a good number of people who are just not skilled enough and will waste your time during the trial period.
Actually, one of them does do some embedded programming for me.
Granted, my sample size is small, but traditional hiring practices are completely ineffective, so you shouldn't waste your time thinking you have any influence over the process.
And honestly, I didn't really even know. I never asked them for a resume and they never provided one.
I don't really test the people I hire. I get to know them on a personal level. I meet them at parties, at meetups, conferences, etc. I then ask if they'd like to make a little side money with a freelance gig. We talk about the work, and if the person says they can do it, I give them a task. A real task, it goes right into the project I'm working on. It usually takes about a week or two.
And if they do a good job, I ask them to do more. If they don't, I pay them for the work they did, thank them, and don't mention they were ever interviewing for a more long-term relationship.
This has been so incredibly much better of an experience than the hiring processes I was involved with as a working stiff. Resumes that don't say shit. Stupidly simplistic coding tests. Interviewing anxiety. All of it, a nightmare.
And it didn't seem to make any difference. We would put more and more effort into hiring, to the point I was spending 10 to 15 hours interviewing and reviewing candidates a week, for weeks on end. It was having a detrimental impact on my performance, without any correlated impact on the performance of the people we hired.
So when I went independent, I decided that, if the results are really, actually random, then that means the inputs have no impact. So I threw away the interviewing process. I just ask people, "can you do this job"?
And really, I don't need to know anything else. That's what it all boils down to. They can lie on their resume just as much as they can lie to the question. They can cram on interview questions the night before. They can be really good at regurgitating academic theory but complete crap at thinking creatively. So I just ask, "can you do this job?"
And then I let them prove it.