Live data from Hacker News

Dear Startups: stop asking me math puzzles to figure out if I can code

countaleph.wordpress.com

161–170 of 354 posts

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#161
My observation is that a lot of interviews come down to "stump the chump" questions; a question that is meant to show a single issue the interviewee has under their belt and is used to gauge the entirety of the interviewer's ability. Math puzzles/logic puzzles are in the same category: they require domain knowledge that probably doesn't translate to any job I've ever worked on.

That aside, one must have a way to measure the abilities of a candidate -- and asking the same set of questions to many people allows you to compare the answers as apples to apples.

I generally don't restrict my people from asking any particular question, but I will ask them to consider what a failed answer really means for the specific job (questions are generally adjusted then).

As an aside, some questions of mine that aren't specifically about coding:

* do you code outside of work (a love of coding translates to good coders)

* send me a link to some code you've written that you are proud of (let see what you got)

* tell me about a problem you had where your solution wasn't correct (how have you dealt with failure).

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#162
post #66

Because they don't know you, you don't have a well known name, they don't know what you can and if it is true. E.g. if somebody hire John Carmack (ID Software), nobody will let him do some math test or ask him trivial programming questions. But you are not John Carmack ;-) It is like in every other job: if you are not a rockstar you are nobody.

There is a job at Google to ask legendary-coders coding questions.

It is reasonable.

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#163
post #85

Earlier quoted context omitted.

I'm fine with only hiring people who have done some work outside their workplace. If they're really great at programming, they have the drive to try new things outside work. Indeed, a couple good personal projects is probably the best predictor of skill you could find. (Skill and interest tend to be highly correlated.)

I used to believe that as well. Now I'm not so sure, mainly through my own personal experience. For 5.5 years (2004-2009), I was a core developer and maintainer over at Xfce. Over that span of time I wrote tens of thousands of lines of code, perhaps more. I did this in addition to my job, which morphed from engineering project management to programming tasks that I didn't find very challenging. After I left that job,…

>> For 5.5 years (2004-2009), I was a core developer and maintainer over at Xfce.

>> But I don't feel the desire to work on personal coding projects anymore. The job fulfills my desire to write code, and outside of that I'd rather be, well... outside.

Another thing to remember is that you're 5-10 years older then you were at that point. Life priorities have changed. You might not be a caffeine-fueled 19 year old anymore, instead you're 28 with 2 kids and a mortgage. Or 34 and just have other things that interest you than tweaking Linux or exploring video drivers.

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#164

Here's the test I've used in the past: Before the interview, I ask them to write some code to access an HTTP endpoint that contains exchange rate data (USD, EUR, GBP, JPY etc.) in XML and to parse and load said data into a relational database. Then to build a very simple HTML form based front-end that lets you input a currency and convert it into another currency. I ask them to send me either a link to a repository (…

> "I don't believe in comments because they slow the PHP interpreter down."

For some bizzare reason, a lot of PHP devs obsess over mini benchmarks and tiny speed issues while ignoring massive latency factors (like the DB).

I find it amusing, though, as PHP hasn't been a direct interpreter for years. It compiles to an internal opcode array and interprets that. While its opcode compiler is silly (it doesn't build an AST and directly emits opcodes, which causes all sorts of things not to work), it does have a proper lexer and I think comments are completely ignored aside from doc comments[1] which can be obtained in userland with reflection.

[1] They look like this:

  /** DocComment */
  /* Normal comment */

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#165

There are many discussions here on HN about company hiring procedures. Company hiring procedures and their effectiveness is a heavily researched topic, but most hiring managers and most job applicants haven't looked up much of the research. After reading the blog post kindly submitted here and some of its comments, and then reading most of the comments here on HN that came in while I was asleep in my time zone, it lo…

I really don't see how this is relevant to this post, since the research you're citing is about hiring people for routinized or semi-routinized labor, like at a car manufacturing plant.

C.f. https://news.ycombinator.com/item?id=4614430

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#167

Here's the test I've used in the past: Before the interview, I ask them to write some code to access an HTTP endpoint that contains exchange rate data (USD, EUR, GBP, JPY etc.) in XML and to parse and load said data into a relational database. Then to build a very simple HTML form based front-end that lets you input a currency and convert it into another currency. I ask them to send me either a link to a repository (…

I really like this. I'm finishing graduate school and I can't tell you how disinterested I am in a company when I apply or send an email and they want to ask me math problems or puzzles....And sadly, more ask puzzles than don't

You wanted a ______ and you are interviewing for a puzzle-solving mathematician really doesn't sound right.

Good stuff.

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#168

Here's the test I've used in the past: Before the interview, I ask them to write some code to access an HTTP endpoint that contains exchange rate data (USD, EUR, GBP, JPY etc.) in XML and to parse and load said data into a relational database. Then to build a very simple HTML form based front-end that lets you input a currency and convert it into another currency. I ask them to send me either a link to a repository (…

I love your interview style: actually test for things the dev will actually do.

I've applied to other jobs who's owners frequent HN, and they did things like test my ability to write sort algos or data structures... but the job called for a good knowledge of JavaScript, Backbone and general knowledge of django to build a mix between a CMS and an elaborate voting system.

They didn't hire me. I'm utterly trash at working on algos under pressure. I felt like a lost child just trying to write a simple sort during the interview. Worse, I made it harder on myself by asking to not write the code in MSWord (or some other similar thing) and instead use jsfiddle. The owner looked at the code and said 'Yeah, that looks about right to me, it should work.' I foolishly hit run.

What I learnt was two-fold:

1) this doesn't test my ability to do the thing you want to pay me to do (as proof, I'm writing more complex applications today at a job I enjoy far more than I would have at that company).

2) The employer can't differentiate between a "smart developer" or "average developer" until the script executes.

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#169
If you're running a startup, the most important thing to hire for is fit. Do they fit in your culture? Do they fit a need that will help you achieve your milestones? Do they fit in the overall growth trajectory of your company? Do they have competency in the specific area you are hiring for and/or where your startup is building overall competency? Can they manage themselves and their time well?

The likelihood of failure of a startup approaches 100%, so you should optimize for likelihood of survival, not for IQ.

If you're not a startup, then the top ranked comment applies. But it doesn't really otherwise.

Re: Dear Startups: stop asking me math puzzles to figure out if I can code

#170

There are many discussions here on HN about company hiring procedures. Company hiring procedures and their effectiveness is a heavily researched topic, but most hiring managers and most job applicants haven't looked up much of the research. After reading the blog post kindly submitted here and some of its comments, and then reading most of the comments here on HN that came in while I was asleep in my time zone, it lo…

Could someone explain what 'mean validity' means in [1]?

Also, how good of a GMA score is needed? That is, how good of a score would be needed for me to stop looking at other candidates?

Post reply on HN