Live data from Hacker News

Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

news.ycombinator.com

1–10 of 46 posts

Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#1
I know that whiteboard coding exercises are in low regard these days, but as someone who's currently looking for a job, I can say they're still widely used. I personally never had a problem with being asked to write code on a whiteboard as part of an interview. Partly because it used to be understood that the point of the exercise was not to write runnable and correct code, rather it was simply to convince the interviewer that under ordinary circumstances (access to an actual computer, sufficient time) that you could produce code that was both runnable and correct.

However, the last three whiteboard coding exercises I've done (two at Facebook and one at a small startup) seemed to be judged much more harshly than what I'm used too. I don't think I "aced" any of them, but had I been interviewing somebody that performed at the same level, I would have said "good enough" and passed the person along for further consideration.

Now some extra information: All three interviews were in Seattle, two of the interviewers for sure were ex-Amazon people (which I mention because I think it might be relevant), and I managed to get hired at Microsoft back in the 90s and at Google in the mid-2000s. I was not hired at Amazon back in 2002-ish, however.

So anyway, I'm wondering: Have I just not been doing as well as these interviews as I think, or has the industry moved the goal posts on me?

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#3
Whiteboard or not, in any interview, you are judged relative to performance of other candidates. At very coveted places like FB, they don't have a dearth of candidates. i.e. Even if you did "good enough", very likely they'll find someone who "aced" the same question on the whiteboard. (We're not debating whether that's right judgment or not, but from their perspective, it's justifiable).

Having said that, I believe that your hunch is also correct viz. tech interviews for core roles at good companies these days, are more competitive than ever. Compared to say 10-15 years ago, the variety of interview problems is higher, complexity of software has increased, and the value of an engineer is understood to be higher ($Millions in equity). All that is contributing to moving of the goal post.

(Source: several years of doing this, now running an interview training course for a living http://InterviewKickstart.com).

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#4
post #2

Despite the sentiments of "they just want to see how you think" or "you don't have to get it right", in my experience if you don't do fairly well on the whiteboard you've failed the interview.

I think this partly depends on how hard the coding question is. If they give you FizzBuzz, I think you're pretty much required to ace it, even if you've never seen it. FizzBuzz is really simple though.

At Facebook I was getting I guess what you'd call classic algorithm questions but with an extra twist thrown in. If you combine that twist with excessive time restrictions (the Facebook interviews were strictly 45 minutes), you not only have to code well, you have to do it fast. On an algorithm that they appear to be hoping you've never seen before.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#5
post #3

Whiteboard or not, in any interview, you are judged relative to performance of other candidates. At very coveted places like FB, they don't have a dearth of candidates. i.e. Even if you did "good enough", very likely they'll find someone who "aced" the same question on the whiteboard. (We're not debating whether that's right judgment or not, but from their perspective, it's justifiable). Having said that, I believe t…

the value of an engineer is understood to be higher ($Millions in equity)

This would be at hip, happening startups in which you take a gamble on awful pay and terrible hours in the hope of being the unicorn, is it? Because in the vast majority of coding jobs, you're not worth millions.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#6
post #4
post #2

Despite the sentiments of "they just want to see how you think" or "you don't have to get it right", in my experience if you don't do fairly well on the whiteboard you've failed the interview.

I think this partly depends on how hard the coding question is. If they give you FizzBuzz, I think you're pretty much required to ace it, even if you've never seen it. FizzBuzz is really simple though. At Facebook I was getting I guess what you'd call classic algorithm questions but with an extra twist thrown in. If you combine that twist with excessive time restrictions (the Facebook interviews were strictly 45 minu…

Yeah, that hope from them is baseless. Truth is, you're competing with candidates who have practiced a lot. They know it, and candidates know it too.

I'll re-iterate: Technical interviews are a competition. You cannot go unprepared. Testing your "raw" skill is stupid and a myth. It's as stupid as saying you want to see Usain Bolt (and everyone else) compete without practice. Those days are gone.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#7
post #3

Whiteboard or not, in any interview, you are judged relative to performance of other candidates. At very coveted places like FB, they don't have a dearth of candidates. i.e. Even if you did "good enough", very likely they'll find someone who "aced" the same question on the whiteboard. (We're not debating whether that's right judgment or not, but from their perspective, it's justifiable). Having said that, I believe t…

the value of an engineer is understood to be higher ($Millions in equity) This would be at hip, happening startups in which you take a gamble on awful pay and terrible hours in the hope of being the unicorn, is it? Because in the vast majority of coding jobs, you're not worth millions.

Value is always relative to supply and demand. Case in point: elementary school teachers are far more valuable, but they are still paid peanuts because there are so many of them.

Engineers who can work on large scale systems in the backend and frontend are not that many. And hence they are valued higher. (I'm not talking about commoditized skills like making webpages and configuring systems, which are indeed vast majority of software jobs).

How many hours you spend is a very good question, but it's a separate question from supply and demand (aka value) of core engineers.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#8
It really depends on the position and the problem. For entry development positions we will give simple tasks that can be 'aced' with any meaningful understanding of (web) development. Oftentimes more complex whiteboard exercises aren't even completed(they could take a long time, but used similarly to ensure problems are approached in an experienced manner.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#9
For what it's worth, I think whiteboard coding exercises can be quite useful, if they are done right. And a complete waste of time if not.

Here "done right" means something like: used as a way to explore problem solving and logic with the candidate.

So in this light, I wouldn't expect anyone to "ace" a problem, but I would expect them to be able to say sensible things about whatever they try and to understand and discuss how to correct issues or alternatives if they are pointed out.

Re: Ask HN: Do you expect interviewees to “ace” whiteboard coding exercises?

#10
I don't expect interviewees to "ace" whiteboard coding exercises, because the point of my questions isn't to answer the question: it's to see how you solve problems. Do you ask smart questions before writing anything? Do you put something down and then refine it? Do you handle errors? Do you understand the performance, maintenance, and testability implications of what you've written? That's what I'm looking for you to answer.

Then again, I work at Pandora, so this is probably an outdated approach...

Post reply on HN