3 ms·

> What 2 hour project can you assign that would let the candidate go into enough depth for you to judge their code, design decisions, architecture skills? Well

> What 2 hour project can you assign that would let the candidate go into enough depth for you to judge their code, design decisions, architecture skills? Well, the movie project suggested in the article was a pretty reasonable one.

silverbax88 10y ago
If you write that in 2 hours, honestly I can't see that being code I'd hire someone for. You'd have to hack together garbage code and you'd look like someone who threw things together without thinking.
- nilkn 10y ago
  Alternatively, with that time limit it's just going to select for candidates who've been churning out Django/Rails/etc. webpages recently and have all the details committed to memory. That feels like premature optimization to me. I can see how it'd find generally good web developers, but I can also see how it'd produce just as many false negatives as Google's interviews do. So is it really that much of a step forward?
  - silverbax88 10y ago
    I completely agree with that. I was thinking, 'yeah, you could hand this to a kid who only knows surface UI and he could crank this out using Rails or something, but it would just be the same garbage that you couldn't maintain later and it wouldn't prove much more than he knows how to write simple apps.' But then maybe that's all they want? I dunno...I've worked with a lot of people who could code a simple app but completely fail when it comes to serious programming.
- clifanatic 10y ago
  Hm - fair point. If I was to do something like this, I'd definitely treat it as a throwaway - I'd do some cursory data modeling, put some tables into a local MySQL DB, reverse engineer that with (say) Hibernate and then throw some Jersey services and some Ajax in front of it. I wouldn't give any consideration to scalability, or security, or maintainability... one would hope that a hiring committee would take that into account, but you're probably right, the people doing the evaluating would likely immediately forget that this was supposed to be thrown together in a couple of hours, people being people as they tend to be.

silverbax88 10y ago

If you write that in 2 hours, honestly I can't see that being code I'd hire someone for. You'd have to hack together garbage code and you'd look like someone who threw things together without thinking.

nilkn 10y ago
Alternatively, with that time limit it's just going to select for candidates who've been churning out Django/Rails/etc. webpages recently and have all the details committed to memory. That feels like premature optimization to me. I can see how it'd find generally good web developers, but I can also see how it'd produce just as many false negatives as Google's interviews do. So is it really that much of a step forward?
- silverbax88 10y ago
  I completely agree with that. I was thinking, 'yeah, you could hand this to a kid who only knows surface UI and he could crank this out using Rails or something, but it would just be the same garbage that you couldn't maintain later and it wouldn't prove much more than he knows how to write simple apps.' But then maybe that's all they want? I dunno...I've worked with a lot of people who could code a simple app but completely fail when it comes to serious programming.
clifanatic 10y ago
Hm - fair point. If I was to do something like this, I'd definitely treat it as a throwaway - I'd do some cursory data modeling, put some tables into a local MySQL DB, reverse engineer that with (say) Hibernate and then throw some Jersey services and some Ajax in front of it. I wouldn't give any consideration to scalability, or security, or maintainability... one would hope that a hiring committee would take that into account, but you're probably right, the people doing the evaluating would likely immediately forget that this was supposed to be thrown together in a couple of hours, people being people as they tend to be.