Live data from Hacker News

My Google Interview Experience

news.ycombinator.com

61–70 of 88 posts

Re: My Google Interview Experience

#61

Not sure if this helps, but as an interviewer I found "masters in computer science from a within top 50 state university in US" to be very weak signal of programming skills. Many candidates from schools which I would consider to be top 20 (like USC) even could not even get past a simple programming question a la "fizzbuzz" test. I would probably trust 5-10 US schools, but it seems for many American universities Maste…

2 years isn't enough time to learn how to program. Then again, I'm not really sure you can reliably make programmers with a substantial degree of success. In my undergrad CS program, we had 300 freshman at the start, and I graduated with 29 others at the end. For people with a CS undergrad, an MS program is quite good. They shore up on depth and theory that they didn't get while working on their programming chops in undergrad.

Sadly, if you can't already program, you can get a master's in CS (because it's focused on depth and theory instead of programming), it's just completely useless. I have no idea what one can do with a MS-CS, if they can't program. It's honestly a little heartbreaking for me, that people to spend 2 years of their lives wasting it on that.

With all that said, I haven't found any degree (at least from any school I've interviewed applicants for) to be a reliable signal for programming. Even if they went to a really good school, there's a good chance that they spent all their time learning network protocols and low-level mechanisms, and will happily write up a sliding-window implementation for me, but will stare at me blankly when I ask for a simple recursive algorithm. It's just tough to find people that spent time studying and practicing general-purpose computer science.

Re: My Google Interview Experience

#62
post #53
post #3

Attention to detail matters. Spending time looking at the code, and talking about (and explaining) your solutions matters. If an engineer misses details and won't spend time looking for a more optimal solution on a white board, I wouldn't expect them to do it in the day-to-day work of their career either.

It's tragic that engineering interviewers extrapolate a candidate's mistakes on a whiteboard during a tense, time pressured interview situation to the calm day coding at a keyboard during a real day at work. It's almost as if we forget what it's like to be an interview candidate when we're the interviewer.

Tragic? What's the alternative? Should we ignore mistakes made at the whiteboard because someone was under time pressure?

Sure, most software engineers (aye, even Google software engineers) are not under significant pressure on an average day. We get to work in the morning, we sit down at our desks, we hack on our code with our 30" monitors and customized editors, and we go home. It's a nice job, really.

But there are occasions when engineers are under pressure. There are deadlines. There are revenue-impacting bugs that need to be fixed while millions of dollars are at stake. There are segfaults in some new code you just deployed and the SRE team that's getting paged at 3am is threatening to burn your house down if you don't fix it before the weekend. Time pressure is not utterly foreign to a software engineer's life, and it's not unreasonable to do what we can to see how an engineer will function under such pressure. An engineer who can only properly function sitting at a desk with his 30" monitor filled with vim instances and ample time on his clock is lacking some measure of flexibility and capability that we reasonably expect to find in the best engineers.

No, we don't forget what it's like to be an interview candidate. You're just forgetting what the point of our interviewing is.

Re: My Google Interview Experience

#63
post #52
post #33

cletus, eta_carinae, and everyone else already gave great points, but I want to add one more to the pile: Google really cares about culture fit. From the way I understand it, there are two main goals for hiring: 1. Always raising the bar by finding people who are better than the average Googler. 2. Find people who would fit in well and Googlers would want to work with. This adds another dimension to the way you could…

Since the interview process doesn't include any sort of personal question or assessments other than one's ability to memorize algorithms and such, How do you measure "culture fit"?

> Since the interview process doesn't include any sort of personal question or assessments other than one's ability to memorize algorithms and such, How do you measure "culture fit"?

It's very subjective and falls clearly in the category of "You recognize it when you see it".

Typical example of a hiring committee deliberation at Google: "He didn't find the optimal solution but he looked very excited about the problem when he was trying to solve it".

Enthusiasm, interest, being personable, excitement about solving problems, all these things count when you interview, especially at Google.

Re: My Google Interview Experience

#64

"Or it could have been that I might have not paid attention to little details which I thought were not important" I think you are underestimating the importance of this. I have hired people before and this is a critical thing I look for. Attention to detail is very important as a developer, if you don't showcase it while on your best behaviour during an interview, I would be concerned what you are like day to day. I…

It irks me that people think that being able to pay attention to detail in a stressful job interview while standing in front of a whiteboard has any strong correlation with their ability and desire to pay attention to detail while doing real-world programming. If this were true, wouldn't MIT and Stanford and CMU grade you while you code at a whiteboard, instead of grading you on significant programming projects? Wher…

I am not necessarily advocating the whiteboard, but am advocating attention to detail no matter what the situation is. I agree the whiteboard can be stressful, but so can a server outage and needing to get a fix out quickly. I don't want a programmer that generally has good attention to detail unless they are under pressure... under pressure is when you need it most.

Re: My Google Interview Experience

#65

Earlier quoted context omitted.

It irks me that people think that being able to pay attention to detail in a stressful job interview while standing in front of a whiteboard has any strong correlation with their ability and desire to pay attention to detail while doing real-world programming. If this were true, wouldn't MIT and Stanford and CMU grade you while you code at a whiteboard, instead of grading you on significant programming projects? Wher…

I am not necessarily advocating the whiteboard, but am advocating attention to detail no matter what the situation is. I agree the whiteboard can be stressful, but so can a server outage and needing to get a fix out quickly. I don't want a programmer that generally has good attention to detail unless they are under pressure... under pressure is when you need it most.

The idea that every programmer should be able to write runnable programs under extreme stress without any errors is ludicrous. That has nothing to do with "attention to detail". It's like professors who give you a failing grade for making a single sign error in the middle of a complicated three page Laplace transform that is otherwise correct. What they really want is to just humiliate you for the power trip.

The "server outage" scenario is a red herring. The vast majority of programmers don't have to fix server outages under pressure that is anything like that at a job interview. Also, you might try hiring programmers instead who write reliable software to begin with. I've written server software, for instance, that has been running continuously for the last dozen years without ever being touched by a human being since. I was also a sysadmin for a seven years, and had to deal with various emergencies all the time. Never was the pressure anything like trying to code at an interview, and the systems I designed and maintained worked great.

Re: My Google Interview Experience

#66
post #37

Earlier quoted context omitted.

> This is true, but it sure would be helpful if they could give qualitative feedback like "the impression was that you were strong in a and b, but weak in x and y." That doesn't seem like too much to ask Sadly, even this kind of feedback exposes the company on a legal front. There is a lot of precedent coming from rejected candidates who ended up suing the company because of this kind of feedback. The only sane thing…

Given how things work in the US, I guess that might be true. But on what grounds does someone get sued for saying "we don't think you know enough about template metaprogramming"? (that's less frivolous than getting sued for saying "we don't have a position for you at the moment", that is, since you don't need a reason to sue.)

A week later, the candidate discovers that the company instead hired someone else that happens to knows absolutely nothing about template metaprogramming. Except that person is whiter / younger / less married than said candidate.

Re: My Google Interview Experience

#67
Facebook gives interview feedback. It reflects their core value of sharing. It's one of a few ways you can see that their process is thoughtfully and intentionally improved upon their predecessors at Google and Microsoft, and reflects the culture of information sharing that defines Facebook.

The "legal" concerns about feedback are bogus.

Google doesn't make "good/not-good" decisions. They make "hire/no-hire" decisions. Huge difference.

They will call you back in 6 months and invite you to apply again. Just memorize the basic half of the CLRS Introduction to Algorithms book, and practice coding solutions to exercises, and you'll get an offer.

Re: My Google Interview Experience

#68

Not sure if this helps, but as an interviewer I found "masters in computer science from a within top 50 state university in US" to be very weak signal of programming skills. Many candidates from schools which I would consider to be top 20 (like USC) even could not even get past a simple programming question a la "fizzbuzz" test. I would probably trust 5-10 US schools, but it seems for many American universities Maste…

Absolutely true. As strange as it may seem, a (terminal) masters degree is almost always less indicative of programming skill than the undergrad degree. I'd expect to be more impressed by a random UWashington Bachelor with no Masters over a UWashington Masters degree holder with a bachelors from somewhere I didn't know.

Re: My Google Interview Experience

#69

Thank you all for your insightful comments. I do feel a little better and more informed. I will try to write an unofficial summary using what others (mostly the official wordings from Googlers (sigh)) have said, mainly for people in the same boat as me. 1. You don't have to be from an Ivy League university to get recruited by them. Yay. 2. Internal references are very very important for both getting recruited and get…

If you want feedback, ask a smart friend (or a professional service) to mock interview you and give detailed feedback. That's better than what any official employer could do.

Re: My Google Interview Experience

#70

I've been through a couple of Google interviews. My last one, I didn't even apply for, they contacted me a couple of months ago, asking if I wanted to interview. Why the heck not. I guess my previous interview was good enough such that they didn't require a phone interview, so I skipped that. Although it's perfectly reasonable to think that they were just trying to hurry the process up for everyone so that's why they…

Try again next year. Consider the math (not exactly how it works, but you'll get the idea):

A job offer depends on 5 flawlessly decided "hire" votes. A mere 5% of interviews are flawed due to interviewer weaknesses (poor training, bad attitude, bad day) or miscommunications. What are the chances that a desirable candidate gets an incorrectly decided "no-offer" result?

Post reply on HN