Dunno, I’ve conducted interviews where it was clear someone had encountered the problem before. There’s some amount of acting skill required to pull off making it seem like you haven’t seen the problem before and “working” it in a way that seems like a way people actually engage with problems and articulate their thought process.
So not only did those people strike me as slightly dishonest, they also made themselves look kind of dumb by being unable to articulate how they were approaching the problem.
Yeah I mean for sure you have to sell it but if you have recently worked through something like a LC medium its just a massively huge ROI to fake it to get through the interview and get an offer. To abuse a double negative LC mediums are not asked by companies that don’t pay well.
So what if you decide to bluff and claim you’ve seen the problem when you haven’t, because you have no idea how to do it and you just want a different question. How often do they call?
Love it but if I was giving the interview and had the misfortune of having to ask LC mediums and some dude/gal said that to me I’d almost certainly say “cool since you know the solution walk me through it rq and we’ll move on” yeah.
God I suck at leetcode. I don’t even have like a starting point approach a ton of these mediums. I’ve been doing googling and like everyone says it just takes time (grinding). Imagine a surgeon having to do a appendectomy for free while being stared at to get a job.
i was saying almost the exact same thing to someone earlier. i’ve come to the conclusion the only way i can improve is by working on “real” projects that focus on whatever i want to improve on, and the more passionate i am about it the better. it doesn’t flex the algorithm puzzle muscles as much, but solving these puzzles was not usually a huge challenge for me, I struggle mightily with the code part.
i dont know why they present so many hard algorithm type problems in interviews, I have yet to see a real world problem that really required this deep type of thinking.
People do it because that’s what’s fashionable at the moment. Back in the day it used to be these bullshit open-ended “puzzle questions” that supposedly provided clues as to the lateral-thinking capability of applicants. There’s an interesting book about this from 2004 called How Would You Move Mt. Fuji?.
People just do whatever industry leaders do out of a cargo-cult mentality because there is no way to separate good applicants from bad in an interview. Some years ago, Google were keeping records of their interview process. When they would hire someone, in addition to approving the hire, the interviewers would predict how well the applicant would subsequently perform in their jobs. They found that the correlation between predicted and actual performance was zero. This led them to overhaul and simplify their interview process.
The good news is it just gets harder the older you get. You know you’re already on strike two when you walk in the door. You know just muddling through isn’t good enough for someone who should clearly be senior. You better knock this thing out of the park and do it faster than everyone else.
yea on some level i realize that. i may have written about it before, but i flamed out of what appeared to be a practically guaranteed job offer in an interview with the CEO that I was assured would be introductory and non-technical and he asked me right out the gate “how would you go about figuring out how many ping pong balls could fit in a 747?” to which I burst out in laughter because I thought he was kidding. he got immediately angry and said he was serious, to which I replied “I wouldn’t.” then, he was like “pretend this interview depends on it” and then I was annoyed so I walked through it.
ended up not getting the job, obviously, but knowing what I know now I was very obviously spared.
And yet if you hire someone you pretty much know in the first week or two (with rare exceptions) if they’re worth a crap or not. Seems like there has to be some way to work that into a hiring process.
On my last round I’m sitting there in these interviews thinking, “if you idiots would just hire me for a week you’d realize what you have.”
lol that isn’t even a lateral thinking problem, it just reduces to “how would you determine the internal volume of a 747” to which the answer is like “look at the specs or talk to Boeing”.
I wonder if the puzzle question fell out of favour in part because the only reasonable answer to any of them now is “I would google that shit”.
I think the problem is that given that programming changes so fast, the major skill you’re looking for is ability to learn quickly, adapt, grasp new concepts, fit into a team and that those things are impossible to assess in an interview when people can grind 500 hours of leetcode to prepare.
The old-school puzzle problem interviews at least recognised the problem of the looks-good-on-paper applicant who is bad at the above, even if their solution was ineffective.
funny enough I did throw these out there to which he said “that isnt helpful for showing me how you think” to which all I could say was something like “this is how I think” and it looked like he just wanted me to know how to calculate the volume of something, which I gave him. lol. I was probably doomed the moment I laughed.
god i need to remember that company’s name so badly right now and i cant. i want to see where their product is at now, the description of their problem sounded insanely stupid to me even at that time, and my skills have grown considerably since.
frequent theme in more recent interviews which I’ve completely DGAF’d my way through has been: engineering teams/principals love me and the moment I talk to a decision maker manager type person it goes down in flames. I need to work on that.
Just estimate like 70 rows, 6 wide, figure out a seat width and distance between seats longitudinally, add in a couple of aisles. The numbers don’t matter at all.
So now we have diameter and length of a tube. Volume of a cylinder is length*pi*r2 (I think). Figure out if he wants ping pong balls in the cargo area or not. And add a fudge factor for the little upstairs part that makes a bump on the plane.
Convert to cubic inches. Say you think a ping pong ball takes up 1 cubic inch. Done.
yea of course which is what I gave him almost exactly that, he even tried to throw little gotchas now and then which I dismantled, but the crux of the issue I sensed was I did not show proper fealty to super important mister c level executive which I am probably at this point completely incapable of doing anymore. I’m very seriously considering pursuing a startup idea I’ve had lately - I have the connections, I think, and I think I know enough about it and how to make a proper investor deck that I could generate at least enough seed funding to get out of this stupid corporate environment for a while, I completely hate it.
It’s such a spectacularly bad example though because the internal dimensions of a 747 are publicly available information and anyone who is like “fuck it I’ll just make wild-ass guesses and multiply numbers together, YOLO” is an idiot.
Like you’re saying though, the actual way the merit of an answer to the question is judged is by how well it communicates to the interviewer “this Big Boy Interviewing Technique that you read about is indeed very clever, thank you for the question, you are very smart”.
I was part of a technical interview for another team last week where the interview lead completely winged it. He didn’t do any prep with anyone else beforehand and in the interview it’s obvious he’s completely winging the interview.
It gets to his portion of the interview and he goes on this spiel about how the position is solving problems where you might not immediately know the answer and then he asks him one of those lateral thinking question but he’s trying to remember it so he phrases it in a way that’s incomprehensible.
The interviewee looks confused and asks him if he could repeat his question.
The guy repeats it but mangles it in a different way so now everyone’s confused. After a bit of this the guy gives up and says let’s move on and then says
“say you run a query and you get a very basic error message(I can’t remember the error message now just that it was so basic that it was obvious he thought of the first thing that came to mind), what would you do to find out what’s wrong with your query?”
The poor interviewee looks even more annoyed and confused because there’s an obvious answer but it’s so basic it can’t be a part of an interview so he matter of factly but hesitantly says
“I’d google it”
The lead, seeing that his question hasn’t opened the interviewee’s mind to us, squints at him and says
“But what would you type in Google?”
The interviewee who’s now both even more annoyed and confused says matter of factly
"I would google, “what is very basic error?”
I had to turn off my camera and mic because I was laughing so hard.