Hacker News new | ask | show | jobs
by happy_path 1567 days ago
I don't think that take-at-home tests are fair. Most of times I've been given one the 3-4 hours have been actually a weekend. In fact, I have the policy of only accepting take-at-home tests that are interesting problems that help me learn something along the process of developing them.

On the other hand, I also don't like on-site interviews. A random question not-related at all with the position you're interviewing for is given, and have to solve it while you're beholded by a bunch of strangers... Being tired/nervous after a 3/4 hour in-site interview, scenic panic... To sum up, tThe recipe for disaster.

Whe I was the interviewer long ago, I gave them a questionary that could be solved in 1 hour. I made sure to tell the candidates that don't get stressed, look for whatever info you want in internet, and that there's no good/bad answers. In fact, I gave them the questionary and once the time has passed I would get them to check the answers.

There was several level of questions, to prune candidates at different levels. That also helped to determine the level of the candidate (I was working for a 200 employee non-tech company at the time).

I suppose that the questionary could have been asked to be completed in-place, in an empty room and without internet... But, what's the point of that? In the real world a developer is allowed to use internet, in fact, knowing to search is a skill I'd like to check.

Besides, if they had some open-source projects I would take a look at that, and at the end, there was a technical conversation between them and some senior developers about their career.

2 comments

I don't feel that take-home tests are fair because of the time that is required, but on the other side of it, I excel at them and have gotten my best jobs from returning excellent results on take-home projects in a short amount of time. For some reason, the format just works better for me and it takes a lot of the pressure out of the process. The downsides become more apparent when you're interviewing for several jobs at the same time and have more than one take-home assignment.
I think lots of companies compensate for take-home tests (as we are doing now). Also I think the willingness to do a take-home is in itself a good filter (but you gotta make it interesting, as parent said)
That's right. That's the main point in favor of making the candidate do a simple questionnaire in office (or remote but in real time). Only 1 hour of their time is wasted.
I think you mean that poorly designed take-home tests aren't fair. Do you think a well-designed, hourlong one could be fair?
I'm genuely interested, what kind of test could that be? Some tests they have gave me in several interviews:

  - Online shops.
  - Hackerrank questions.
  - Read from a big log file in an efficient manner with Python (did it but there was also a 3-4 hour on-site interview).
Any test that: 1) is similar to the work in the day-to-day role 2) doesn't require more than a 1-2 hour time commitment from a candidate 3) is interesting

HackerRank questions typically don't meet criteria 1 or 3. The best tests are often ones that the engineering team has designed themselves.

For example, this is an example of a test that meets all the criteria above: https://app.mightyacorn.io/flow-club/founding-front-end-engi...