15 ms·

The current state of job interviews is terrible. Just the fact that studying for a few weeks affects their outcome should raise eyebrows everywhere. And don't

by oDot 6y ago

The current state of job interviews is terrible. Just the fact that studying for a few weeks affects their outcome should raise eyebrows everywhere. And don't get me started on lack of job-related questions, false negatives and having to guess what interviewers want to hear

kinkrtyavimoodh 6y ago
[Edit: Parent commenter has now reworded their comment to remove confusion] Huh? OP wasn't a rando who just 'studied' for a few weeks and passed. He has been working in Tech for 7+ years [1]. If anything the moral of this story is it doesn't matter that you have a 7 year experience, you still have to do fucking Leetcode for 2 months if you want to get a new job. [1] https://www.linkedin.com/in/bharathpbhat/ https://www.linkedin.com/in/bharathpbhat/
- oDot 6y ago
  > ... it doesn't matter that you have a 7 year experience, you still have to do fucking Leetcode for 2 months if you want to get a new job Exactly my point -- two weeks of leetcode do not significantly affect the candidate's ability to perform the job, but it has immense influence on how well s/he'll do in the interview
  - kinkrtyavimoodh 6y ago
    If that was your point I think you didn't state it correctly. Your original comment makes it sound like you think OP undeservedly got a job merely be hacking through a couple weeks of interview prep.
    - oDot 6y ago
      Edited for what I hope is clarity ;p
- ramraj07 6y ago
  I see _some_ potential acceptability to this idea though. There are a few hypotheses for this to be true: 1. If you're not a good programmer even with leetcode practice you can't get good 2. If you're a good programmer with leetcode practice you eliminate anxiety and are able to perform optimally according to your programming skills Thus, it is definitely semi annoying that you need to "prep" for interviews with leetcode, its not the absolute worst thing in the world. Software engineers in the US tend to routinely forget how good they have basically everything compared to every other profession / nation out there. This is also coming from and Indian dude currently back in India after a decade of being in the States. I think a 60 day window is perfectly acceptable (not great but not terrible either) for immigrant workers. I'm still glad for the opportunity this gives people. You're supposed to be a smart cookie anyways right? Didn't you know these terms when you started? Perhaps you should have made sure you had an exit plan. It's definitely on you if you come on a h1b and instantly proceed to buy a mansion that you can't sell off and leave nation if needed. Also, people blame their kids for not wanting to move them back to their original countries - I feel like the kids will be alright. It's probably the adults who will despise going back to a "lesser" lifestyle or something.
  - bradlys 6y ago
    Being a good programmer doesn't necessarily mean you're good at leetcode and just need to brush up. I don't meet almost anyone who uses DP in their daily life but you can bet your ass you'll run into a problem that requires DP in a FAANG interview. And you better put it on the board in under 20 minutes or you're getting passed for the next guy who did 1000+ problems as their prep. It wouldn't be so bad if being a good programmer was sufficient and you just needed a little bit of brushing up - but the arms race started years ago... So, now the bar is very high. Leetcode is solving small problems, often with tricks, in a very short period of time. Being a software engineer is solving large problems over large periods of time.
    - stack_underflow 6y ago
      It's interesting, when I tell people they need to be prepared for this bar I get a surprising number of people who just don't seem to want to accept that it's true or flat out claim that it's not true when I know they haven't interviewed in years. In the past 2 years I've done about 40ish+ interviews at all the big tech co's + smaller-midsize startups and aside from 3 or 4 take home assignments I was given, _every_ interview was leetcode style, to the point where I would just start noting which exact leetcode questions I was asked when friends asked about my experience. The only exception would be 1 or 2 "system design" rounds out of the 5/6 on-site interviews if the company was calibrating me for senior level.
lordnacho 6y ago
It's bad, but compared to other industries there are some real pluses. At least you can actually hunker down and do these silly tests. There's a huge amount of advice on the technical stuff. And yes it means people who have a job already are disadvantaged, but that also means people who have no income have a way in. In a lot of other industries, there's no real way to boost your game. Often it's purely reputation of your school and previous employers, and your feedback on a failed application is just platitudes. People circle around for ages and never find anything but crappy advice that doesn't work. It's also a process that makes bias a little bit harder to introduce. If your pal from the country club can't fizzbuzz, it will be hard to convince people to accept him. Not saying this is bullet proof, but in most jobs it's a lot more vague whether someone is terrible.
- gcheong 6y ago
  If only it had stopped at fizzbuzz as a quick check for whether you can meet the lowest bar for coding I think it would still have some merit but it just seems to have gotten out of hand to the point that ability to get a job now is completely decoupled from your prior experience.
  - bradlys 6y ago
    It's an arms race. Big name employers only wanted the top 1% of candidates. A bunch of people wanted to get into the top 1% and figured out a way (leetcode, ctci, etc.). And then employers kept only wanting the top 1% - so they just raised the bar higher and higher. Now we're at the scale where if you haven't done a minimum of 200 problems before any interview round - you're not likely getting an offer from FAANG/etc.
    - triyambakam 6y ago
      Ctci?
    - hinkley 6y ago
      I wonder if anyone at FAANG has figured out that they collectively employ over 1% of the developers in the US. What they are trying to do is not just practically impossible, it is literally impossible.
TrackerFF 6y ago
It seems to be a mostly US-centric thing. I live in Scandinavia, and I have yet to encounter the levels of rigor I see in the US (tech) interviewing process - which is kind of weird, as we have much, much more stringent workers rights/laws, making it more difficult to get rid of a bad hire. Sure, US companies, especially in tech hubs, operate on a completely different scale than companies over here - at least in terms of funding/capital, size, and impact; So one might suspect your companies and startups to be more selective. But yet, it seems like a completely different world. Over here, there seems to be much more emphasis (and trust) on your resume, and the hiring process is more focused on fit. If you have the basics down, most can be thought - but it's pretty difficult to teach culture.
- lowiqengineer 6y ago
  > Over here, there seems to be much more emphasis (and trust) on your resume, and the hiring process is more focused on fit. If you have the basics down, most can be thought - but it's pretty difficult to teach culture. That's probably because OP makes triple to quadruple your total compensation at minimum as an L5 ML Engineer at Google. https://www.levels.fyi/ https://www.levels.fyi/ Total: $356,119
  - TrackerFF 6y ago
    Obviously the salaries are quite different, but at the same time - you're pretty much walking around in golden shackles, no? I can't imagine those salaries being relevant, other than a handful of places in the US. And for the price of a small 2-bedroom apartment in the SF-area, you can pretty much buy a mansion where I live. Then you have things like healthcare, school costs, daycare / cost of raising a child, etc. etc.
    - lowiqengineer 6y ago
      Cost of healthcare is minimal at Google. And yeah, you can easily afford a mansion if you want to commute long enough. 300k to 500k is enough for nice housing - unfortunately, as someone making $150k at 24 I’ll never have this American dream.
jp0d 6y ago
It does seem like a US-centric thing. I've faced several interviews here in Australia and the teams care more about how the candidate will fit into the team. Most them sort of assume that what's written in the resume is correct. So the interview process is basically just trying to validate those claims. Also, it's really important to get your references right. But it depends a lot on the companies. Some of them have take home projects and white board problem solving sessions. We don't have anything like the H1B here. Either you're a permanent resident or an Aussie citizen. The only difference being the ability to vote and jury duty. Sure, some of the govt jobs are out of reach for permanent residents. The work visa holders are generally transferred by a company from another country.
omegote 6y ago
> studying for a few weeks affects their outcome should raise eyebrows everywhere Lol then what do you think about the hefty exams people have to go through in Spain to become a public employee? People spent several years just studying after finishing college just to get a job - for life, that is.

by oDot 6y ago

kinkrtyavimoodh 6y ago

[Edit: Parent commenter has now reworded their comment to remove confusion] Huh? OP wasn't a rando who just 'studied' for a few weeks and passed. He has been working in Tech for 7+ years [1]. If anything the moral of this story is it doesn't matter that you have a 7 year experience, you still have to do fucking Leetcode for 2 months if you want to get a new job. [1] https://www.linkedin.com/in/bharathpbhat/ https://www.linkedin.com/in/bharathpbhat/

oDot 6y ago
> ... it doesn't matter that you have a 7 year experience, you still have to do fucking Leetcode for 2 months if you want to get a new job Exactly my point -- two weeks of leetcode do not significantly affect the candidate's ability to perform the job, but it has immense influence on how well s/he'll do in the interview
- kinkrtyavimoodh 6y ago
  If that was your point I think you didn't state it correctly. Your original comment makes it sound like you think OP undeservedly got a job merely be hacking through a couple weeks of interview prep.
  - oDot 6y ago
    Edited for what I hope is clarity ;p
ramraj07 6y ago
I see _some_ potential acceptability to this idea though. There are a few hypotheses for this to be true: 1. If you're not a good programmer even with leetcode practice you can't get good 2. If you're a good programmer with leetcode practice you eliminate anxiety and are able to perform optimally according to your programming skills Thus, it is definitely semi annoying that you need to "prep" for interviews with leetcode, its not the absolute worst thing in the world. Software engineers in the US tend to routinely forget how good they have basically everything compared to every other profession / nation out there. This is also coming from and Indian dude currently back in India after a decade of being in the States. I think a 60 day window is perfectly acceptable (not great but not terrible either) for immigrant workers. I'm still glad for the opportunity this gives people. You're supposed to be a smart cookie anyways right? Didn't you know these terms when you started? Perhaps you should have made sure you had an exit plan. It's definitely on you if you come on a h1b and instantly proceed to buy a mansion that you can't sell off and leave nation if needed. Also, people blame their kids for not wanting to move them back to their original countries - I feel like the kids will be alright. It's probably the adults who will despise going back to a "lesser" lifestyle or something.
- bradlys 6y ago
  Being a good programmer doesn't necessarily mean you're good at leetcode and just need to brush up. I don't meet almost anyone who uses DP in their daily life but you can bet your ass you'll run into a problem that requires DP in a FAANG interview. And you better put it on the board in under 20 minutes or you're getting passed for the next guy who did 1000+ problems as their prep. It wouldn't be so bad if being a good programmer was sufficient and you just needed a little bit of brushing up - but the arms race started years ago... So, now the bar is very high. Leetcode is solving small problems, often with tricks, in a very short period of time. Being a software engineer is solving large problems over large periods of time.
  - stack_underflow 6y ago
    It's interesting, when I tell people they need to be prepared for this bar I get a surprising number of people who just don't seem to want to accept that it's true or flat out claim that it's not true when I know they haven't interviewed in years. In the past 2 years I've done about 40ish+ interviews at all the big tech co's + smaller-midsize startups and aside from 3 or 4 take home assignments I was given, _every_ interview was leetcode style, to the point where I would just start noting which exact leetcode questions I was asked when friends asked about my experience. The only exception would be 1 or 2 "system design" rounds out of the 5/6 on-site interviews if the company was calibrating me for senior level.

lordnacho 6y ago

It's bad, but compared to other industries there are some real pluses. At least you can actually hunker down and do these silly tests. There's a huge amount of advice on the technical stuff. And yes it means people who have a job already are disadvantaged, but that also means people who have no income have a way in. In a lot of other industries, there's no real way to boost your game. Often it's purely reputation of your school and previous employers, and your feedback on a failed application is just platitudes. People circle around for ages and never find anything but crappy advice that doesn't work. It's also a process that makes bias a little bit harder to introduce. If your pal from the country club can't fizzbuzz, it will be hard to convince people to accept him. Not saying this is bullet proof, but in most jobs it's a lot more vague whether someone is terrible.

gcheong 6y ago
If only it had stopped at fizzbuzz as a quick check for whether you can meet the lowest bar for coding I think it would still have some merit but it just seems to have gotten out of hand to the point that ability to get a job now is completely decoupled from your prior experience.
- bradlys 6y ago
  It's an arms race. Big name employers only wanted the top 1% of candidates. A bunch of people wanted to get into the top 1% and figured out a way (leetcode, ctci, etc.). And then employers kept only wanting the top 1% - so they just raised the bar higher and higher. Now we're at the scale where if you haven't done a minimum of 200 problems before any interview round - you're not likely getting an offer from FAANG/etc.
  - triyambakam 6y ago
    Ctci?
  - hinkley 6y ago
    I wonder if anyone at FAANG has figured out that they collectively employ over 1% of the developers in the US. What they are trying to do is not just practically impossible, it is literally impossible.

TrackerFF 6y ago

It seems to be a mostly US-centric thing. I live in Scandinavia, and I have yet to encounter the levels of rigor I see in the US (tech) interviewing process - which is kind of weird, as we have much, much more stringent workers rights/laws, making it more difficult to get rid of a bad hire. Sure, US companies, especially in tech hubs, operate on a completely different scale than companies over here - at least in terms of funding/capital, size, and impact; So one might suspect your companies and startups to be more selective. But yet, it seems like a completely different world. Over here, there seems to be much more emphasis (and trust) on your resume, and the hiring process is more focused on fit. If you have the basics down, most can be thought - but it's pretty difficult to teach culture.

lowiqengineer 6y ago
> Over here, there seems to be much more emphasis (and trust) on your resume, and the hiring process is more focused on fit. If you have the basics down, most can be thought - but it's pretty difficult to teach culture. That's probably because OP makes triple to quadruple your total compensation at minimum as an L5 ML Engineer at Google. https://www.levels.fyi/ https://www.levels.fyi/ Total: $356,119
- TrackerFF 6y ago
  Obviously the salaries are quite different, but at the same time - you're pretty much walking around in golden shackles, no? I can't imagine those salaries being relevant, other than a handful of places in the US. And for the price of a small 2-bedroom apartment in the SF-area, you can pretty much buy a mansion where I live. Then you have things like healthcare, school costs, daycare / cost of raising a child, etc. etc.
  - lowiqengineer 6y ago
    Cost of healthcare is minimal at Google. And yeah, you can easily afford a mansion if you want to commute long enough. 300k to 500k is enough for nice housing - unfortunately, as someone making $150k at 24 I’ll never have this American dream.

jp0d 6y ago

It does seem like a US-centric thing. I've faced several interviews here in Australia and the teams care more about how the candidate will fit into the team. Most them sort of assume that what's written in the resume is correct. So the interview process is basically just trying to validate those claims. Also, it's really important to get your references right. But it depends a lot on the companies. Some of them have take home projects and white board problem solving sessions. We don't have anything like the H1B here. Either you're a permanent resident or an Aussie citizen. The only difference being the ability to vote and jury duty. Sure, some of the govt jobs are out of reach for permanent residents. The work visa holders are generally transferred by a company from another country.

omegote 6y ago

> studying for a few weeks affects their outcome should raise eyebrows everywhere Lol then what do you think about the hefty exams people have to go through in Spain to become a public employee? People spent several years just studying after finishing college just to get a job - for life, that is.