Hacker News Reader

Comment by geebee | Hacker News Reader

You may want to consider these posts by Gayle Laakman McDowell (who write "Cracking the Coding Interview") on possible problems with take-home exams:

http://www.gayle.com/blog/2013/09/18/companies-who-give-cand...

From her blog post on general interviews and take-homes:

http://www.gayle.com/blog/2015/6/10/developer-interviews-are...

"Be fair to candidates. Balance the demand on their time with the utility you get from it. If you're giving a longer project, these should largely replace your interviews. Be aware that this means losing candidates who have other responsibilities."

(in the context of github and other outside contributions, but applicable in my opinion to take home exams as well): "You may really struggle with female developers here. They're more likely to have substantial family responsibilities (and thus can't do as much out of work)."

View on HN

Same experience as OP here both on number of applicants and number of qualified female applicants.

Take home test results were considerably worse for females too and they had over five days to complete a simple thirty minute exercise.

There might be responsibilities that get in the way but IMO if you can't find a slice of time to work on something that came up then I'll be really careful in considering you for a job.

Shit happens w/o planning and I'd rather have people I can count on.

I agree, the inability to do homework would itself be a little bit of a red flag for me, beyond the quality of the homework.

It may be that the candidate doesn't want to waste time. There's a big difference between putting in some time on weekends and evenings for an important work issue that arises, vs doing some homework assignment that might not even get looked at.

I think this is one of the reasons Ms McDowell suggest doing the take home project later in the process, only once a candidate is promising, with a higher pass rate.

If you're using it simply to screen (i.e., if your pass rates are low), and the take home demands are substantial, you are likely wasting a huge amount of candidates time, in aggregate. This can legitimately harm a company's reputation among developers.

ISP network engineering/operational support software team here: People need to be able to find two hours of time (over a 5-6 day period) to complete a small take home project. When they do it and how they do it is up to them.

If somebody can't do this then they would not be a good fit to be in the on-call rotation where 1 out of every 5 or 6 weeks you could have alert emails and phone calls ringing on your phone at 3am. Sorry, but five to six nines statistical uptime over one year demand it.

30 minutes sounds less objectionable than what I've experienced.

The episode that turned me badly against take home tests involved a two-phase approach. Both were handled through a recruiter. The first was about a one hour set of questions (what does this code do, implement this class, explain dependency injection, that short of thing). That went well, so they then moved to a homework project they advised to spend no more than 5-7 hours on.

I did this, sent it back in and... crickets chirping. I called the recruiter (in house) every week or so to follow up. In about a month I got the standard "we've decided not to pursue your application further at this time.

I honestly have no idea if anyone even looked at it. I certainly never talked to a developer.

I can see how a 30 minute exercise could be a good thing, in that it could save everyone some time. Even then, though, I see it as a red flag if a company wants too much of my time before I talk to a developer in the recruiting process.

Keep in mind also, there's a huge difference between being available to work on something that comes up in a job, and shaking free 5-7 hours for a take home test that a company only might even bother looking at!

Reply on news.ycombinator.com