4 ms·

This is a good point. Traditional recruiters have different ways of knowing how good you are at programming (often they don't). By starting with your Gridmancer

by nwinter 13y ago

This is a good point. Traditional recruiters have different ways of knowing how good you are at programming (often they don't). By starting with your Gridmancer competency, we can potentially find and qualify you for opportunities that traditional credentials couldn't. (And we're not going to send candidates who aren't good enough to higher-skill positions.) One weakness is that we don't have a good way of measuring how much software development (as opposed to core programming skill) you have. In this sense, we don't yet have an advantage over traditional recruiters. Traditional recruiters do have better connections, for now, but we are going really high-hustle for this initial batch of players, so it's not like we're just sending off resumes.

jschmitz28 13y ago
Is the general consensus now that core programming skills and being good at math are the same thing? Because after taking a look at this challenge it appears to be a math problem that requires mathematical knowledge and barely any programming knowledge.
- darylteo 13y ago
  Glad I'm not the only one who thought this. Exercises like these have a very poor accuracy if it is used it to test for developers.
  - stcredzero 13y ago
    See above. Could be testing for problem solving ability and willingness to try.
- stcredzero 13y ago
  Did you even try it? You can get pretty far by just tinkering. I came up with a O(n) algorithm that gets me 30 rectangles just by playing around. In general, you can get "pretty good" results just by taking low hanging fruit then optimizing.
pmiller2 13y ago
I'm skeptical of this whole approach. I've done a ton of these code challenges (ranging from literally FizzBuzz level up to stuff involving dynamic programming, recursion, and NP-complete problems) on CodeEval, and it's never gotten me anywhere close to a new job (though some of them were kind of fun). What makes you think your code challenges are any better than theirs for this particular purpose?
- hmsimha 13y ago
  I was thinking the exact same thing. I was really excited when I discovered codeEval, but after solving a few of the 'hard' challenges on there it became clear to me that hiring managers were looking for more well-rounded programmers with demonstrated experience in software development (something I've been working on through my own post-academic studies), not someone who can whip out a solid recursive algorithm to solve a brain teaser. And it makes perfect sense to me, the former is more likely to be actually useful in the day-to-day of a software engineering position, and the latter might only come in handy once in a blue-moon, or for more mathematics heavy domain-specific positions that require less code ejaculation and more optimization anyway.
  - pmiller2 13y ago
    It's actually even worse in my case. I do have 1.5 years of actual engineering experience at a real company. That tells me that CodeEval (and maybe "code challenges" in general) have at most very little and at worst negative signal value. (Well, either that or I'm a terrible engineer, but I think they would have fired me long before a year and a half if that were the case.)

by nwinter 13y ago

jschmitz28 13y ago

Is the general consensus now that core programming skills and being good at math are the same thing? Because after taking a look at this challenge it appears to be a math problem that requires mathematical knowledge and barely any programming knowledge.

darylteo 13y ago
Glad I'm not the only one who thought this. Exercises like these have a very poor accuracy if it is used it to test for developers.
- stcredzero 13y ago
  See above. Could be testing for problem solving ability and willingness to try.
stcredzero 13y ago
Did you even try it? You can get pretty far by just tinkering. I came up with a O(n) algorithm that gets me 30 rectangles just by playing around. In general, you can get "pretty good" results just by taking low hanging fruit then optimizing.

pmiller2 13y ago

I'm skeptical of this whole approach. I've done a ton of these code challenges (ranging from literally FizzBuzz level up to stuff involving dynamic programming, recursion, and NP-complete problems) on CodeEval, and it's never gotten me anywhere close to a new job (though some of them were kind of fun). What makes you think your code challenges are any better than theirs for this particular purpose?

hmsimha 13y ago
I was thinking the exact same thing. I was really excited when I discovered codeEval, but after solving a few of the 'hard' challenges on there it became clear to me that hiring managers were looking for more well-rounded programmers with demonstrated experience in software development (something I've been working on through my own post-academic studies), not someone who can whip out a solid recursive algorithm to solve a brain teaser. And it makes perfect sense to me, the former is more likely to be actually useful in the day-to-day of a software engineering position, and the latter might only come in handy once in a blue-moon, or for more mathematics heavy domain-specific positions that require less code ejaculation and more optimization anyway.
- pmiller2 13y ago
  It's actually even worse in my case. I do have 1.5 years of actual engineering experience at a real company. That tells me that CodeEval (and maybe "code challenges" in general) have at most very little and at worst negative signal value. (Well, either that or I'm a terrible engineer, but I think they would have fired me long before a year and a half if that were the case.)