Live data from Hacker News

Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

news.ycombinator.com

191–200 of 206 posts

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#191
post #73

BALL-PITS and CLOWN NOSES I worked with a guy, Bob, in our college computer room. Bob had returned to get his BS after many years in industry. He had great stories. He also had great advice. He taught me that you can learn a lot about how a company will treat their employees by paying attention to the way they treat you during the interview. He suggested several sign that indicated it was best to just "walk away". If…

I'd have to "just walk away" from the entire industry, since I've never once in over twenty years had an interview which did not involve a demonstration of my ability to reason through a problem and communicate a proposed solution using conversation and a whiteboard. I am a professional, yes, and thinking about problems and communicating about solutions is my job; why should I be offended that someone wants me to show them how I work?

Why would I ever want to hire someone incapable of demonstrating their skills in such a simple, straightforward way? If you can't solve a toy problem and explain your solution, or at least have a conversation about the constraints and tradeoffs as you work on it, how am I to imagine there is any chance you could be a productive teammate?

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#192

Earlier quoted context omitted.

Top 1% of intelligence

About 8 years ago, I got an interview at Google. I didn't really even want to work there... I just wanted to see what the place was like. So I made it through the phone screens, went to the interview, and I failed. It was still a good experience!

"I got an interview at Google. I didn't really even want to work there... I just wanted to see what the place was like"

... sure

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#193
post #120

Earlier quoted context omitted.

Top 1% of intelligence

Intelligence is hard to define. There's no scientific, widely-agreed-upon definition. Even if there were, algorithm tests on whiteboards would be a poor test, because they also involve prior knowledge of the task and a very specific type of intelligence. In my experience, social intelligence is a better predictor of success in any workplace, anyway.

"Intelligence is hard to define."

and then you type:

"In my experience, social intelligence is a better predictor of success"

Something tells me you don't work at google :D

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#194
post #183
post #181

Earlier quoted context omitted.

Apparently they added an author since I last studied. Damn, that was over 10 years ago.

Yeah, quite a bit must have changed or more likely evolved through the years, so a lot of the pseudocode and algorithms have been modified or added for instance, topics on dynamic programming, memoization and the like have had more emphasis. Several revisions to examples in basic tree algorithms and additions of some too have arisen. You could consider brushing up on a few things as a casual read perhaps, especially…

That's OK, I don't plan to read any CS books in the future, with the exception of The Art of Computer Programming, which I feel, after many years of training (including a decade as a SWE at Google), I'm finally ready to enjoy.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#195

Earlier quoted context omitted.

About 8 years ago, I got an interview at Google. I didn't really even want to work there... I just wanted to see what the place was like. So I made it through the phone screens, went to the interview, and I failed. It was still a good experience!

"I got an interview at Google. I didn't really even want to work there... I just wanted to see what the place was like" ... sure

I wanted to see if I could get in.

I don't like large corps, period.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#196
post #194
post #183

Earlier quoted context omitted.

Yeah, quite a bit must have changed or more likely evolved through the years, so a lot of the pseudocode and algorithms have been modified or added for instance, topics on dynamic programming, memoization and the like have had more emphasis. Several revisions to examples in basic tree algorithms and additions of some too have arisen. You could consider brushing up on a few things as a casual read perhaps, especially…

That's OK, I don't plan to read any CS books in the future, with the exception of The Art of Computer Programming, which I feel, after many years of training (including a decade as a SWE at Google), I'm finally ready to enjoy.

Oh yes, one of my bucket list kinda goals in life is to be able to get to a level where I can grok TAoCP at a visceral level and understand the underlying mathematics used to derive the algorithmic complexities and how they all function. I think any computer science person who is passionate about the topics should give it a try at some point in their lives.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#197
post #32

Earlier quoted context omitted.

Do you have a cite/example of a lecturer doing that on an exam? Are there any examples where that has been successful? I had never heard of that happening! It would be extremely cool if it ever worked.

Not quite the same, but David Huffman was given a choice between writing a term paper on finding the optimal binary code and a final exam. Just before he was about to give up on the term paper, he solved the term paper problem by inventing Huffman Coding[1]. [1] http://www.huffmancoding.com/my-uncle/scientific-american

my own huffman story: I took huffman's class, cybernetics, in college (thinking that it would teach me how to build robots that interfaced with humans). However, the class was mostly about how to build circuits, and Karnaugh maps weighed heavily. The final problem on the final test was a tricky Karnaugh map and I got it wrong- although I knew that you could form squares that "wrapped" around the top and bottom edges, but he put a square that was wrapped around the four corners: 1 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 and I totally missed that the symmetry applied in both directions at the same time (I'd call that a tricky question rather than a trick question).

That class led to a ton of imposter syndrome that took me decades to overcome. Thanks, Huffman.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#198
post #41

Earlier quoted context omitted.

I got the same response from an old friend who has worked for Google for the better part of a decade. FWIW my understanding is that the Hiring Committee tends to discard feedback from interviewers like your latter type here.

Is there any mechanism like those olympics diving competition? i.e. the highest and lowest score are discarded.

They don't do this as a matter of course, but they do look at the question that was asked various factors surrounding the interview. They certainly can and do throw out interviews that don't generate any signal.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#199
post #73

BALL-PITS and CLOWN NOSES I worked with a guy, Bob, in our college computer room. Bob had returned to get his BS after many years in industry. He had great stories. He also had great advice. He taught me that you can learn a lot about how a company will treat their employees by paying attention to the way they treat you during the interview. He suggested several sign that indicated it was best to just "walk away". If…

I'd have to "just walk away" from the entire industry, since I've never once in over twenty years had an interview which did not involve a demonstration of my ability to reason through a problem and communicate a proposed solution using conversation and a whiteboard. I am a professional, yes, and thinking about problems and communicating about solutions is my job ; why should I be offended that someone wants me to sh…

That's nice in theory. But it doesn't fit the facts.

1) The whiteboard questions are unrelated to the job opening.

The interviewers have no idea why you are there. They don't know what your resume says. They don't know who you might be working for. They don't know what you might be working on. Would you hire someone to work for you without knowing if they were going to clean your house, install new lighting, or rebuild your kitchen? How exactly would you interview a person if you don't know why you need their skills? Or how their skills are related to the task? Asking a carpenter to "whiteboard" the steps to clean a bathtub during an interview is insulting.

2) The interview is being conducted by people who have almost no job experience.

My experience is that the interviewers are recent college graduates. And that the whiteboard questions are related to their data structures class. They are focused on details like worst-case behavior of the algorithm and whether you can quote the log-log-n asymtotic performance. This entirely ignores the fact that the actual behavior depends a lot on the actual measured performance. Take, for example, my paper on garbage collection (http://daly.axiom-developer.org/TimothyDaly_files/Bozm84.pdf) where the actual data access pattern matters. Theoretically strong algorithms usually exhibit bad behavior because of data access patterns, CPU caches, look-aside buffering, and/or branch prediction failures.

3) There is no known relation between being effective and being able to present information.

I know people who are strongly autistic, to the point of not making actual eye contact or giving any verbal feedback at all, appearing to be "not paying attention"... and then create beautifully crafted excellent code. I also know people who "talk a good game" and couldn't write a "hello world" page in html. Please point me to any study that shows that whiteboarding is an effective way to judge candidates. Even Google's HR head claims it is ineffective.

4) There is a ball-pit / clown nose aspect to the interview.

"Everybody does it" so wear the clown nose. I've had a fair number of interviews in my 48 year career. Most of them were with people who needed something done and I could demonstrate my ability to do that. Interviews exist for a reason. Asking me to whiteboard Knuth-Bendix because you're writing an editor is reasonable. But the same question for a customer service job would be nonsense.

These "clown-nose" interviews are simply insulting and a waste of time. Walk away.

Re: Ask HN: Can engineers from Google or Facebook solve whiteboard questions easily?

#200
post #101
post #36

I work at Google and no it's not easy for us either. I've heard many of my coworkers joke that if they went through the interview process again they'd probably fail. That said I see two kinds of interviewers. One kind (the good kind) takes a medium difficulty question and uses it to explore the candidate's coding, algorithms, communication, and problem solving skills. The other kind has a super hard question with a s…

I also work at Google and what mystifies me when I do interviews is just how consistently people manage to screw up even the medium difficulty questions. I used to love asking this question until it was banned: https://programmingpraxis.com/2012/01/20/knights-on-a-keypad... This is a great question because it has several levels of solutions: 1. A recursive solution that iterates over all sequences, exponential time a…

Fun exercise. This is how far I got after about 45 minutes: https://gist.github.com/bjourne/7e45d41e8e4f5518bb92f6df7978... I wasted half the time not getting that you don't need to store the paths only the counts! But this was in my quiet home and I would probably fluke out in an interview setting where you have to talk while you code.

I'm still going to consider myself a pretty good coder. Perhaps not Google material, but pretty damn good still. :)

Post reply on HN