8 ms·

> FAANG are struggling to fill demand in hiring and are offering increasingly high salaries This has been the case, according to my colleagues who are older, s

> FAANG are struggling to fill demand in hiring and are offering increasingly high salaries This has been the case, according to my colleagues who are older, since at least 2003 ;-)

vmception 5y ago
> Recruiters are charging 22-25% for placements, and having difficulties filling them. This has been the same as well, to my knowledge, since 2010
rubicon33 5y ago
And yet, they still persist on presenting leetcode riddles that have very little relevance to the job being interviewed for. If they were hurting that much for devs, you'd think they might lower the bar a little by not requiring a years study of leetcode.
- kamarg 5y ago
  It's more expensive to make a bad hire than to not hire someone and move a bit slower.
  - skeeter2020 5y ago
    while this is true, maybe they should work on how they assess candidates instead of continually complaining about how there are none.
  - kkdaemas 5y ago
    So... they double-down on a poor proxy of effectiveness on the job?
  - lanstin 5y ago
    Leet code is neither sufficient or necessary to writing brilliant software. Your statement is true but Google, being mostly people cut from the same cloth, has group think about these issues. On the other hand, that monolithic culture is probably what has limited their success and kept them from being a more harmful monopoly.
  - mech422 5y ago
    google admitted years ago, it didn't help.. https://www.theatlantic.com/business/archive/2013/06/google-finally-admits-that-its-infamous-brainteasers-were-completely-useless-for-hiring/277053/ https://www.theatlantic.com/business/archive/2013/06/google-...
    - fshbbdssbbgdd 5y ago
      Google found that riddles didn’t help, so they stopped asking them. They didn’t find the same for coding questions.
      - mech422 5y ago
        Yet :-P I wouldn't be surprised to see a similar article about leet code...
  - marcosdumay 5y ago
    There are two problems with this phrase, one is that it shouldn't. Really, if somebody a bit worse goes way into the negative value for you and you are a large company, you have some procedural problem that should be fixed. Long-time employees also have bad moments, and you should be able to survive those. The second problem is that the onus is on you on making sure those bad interviews decrease the odds if hiring bad people, instead of increasing them or being irrelevant. Without that evidence, this is a non-argument.
  - smnplk 5y ago
    Why not train a bad hire ?
    - dagw 5y ago
      Training a 'bad' hire requires taking 'good' employee out of circulation for a while and, depending on what way they are bad, might not lead to anything anyway.
      - dagw 5y ago
        I want to add to this that in many cases getting many "bad hires" can also be symptom of bad management or bad workplace environment.
- xyzelement 5y ago
  If I am running the company, I would MUCH rather operate short-handed than "fix" my recruiting problems by lowering the bar. You don't have to agree with the bar the companies are using, but that is the bar they have converged on, and one that plenty of people are capable of passing. Would much rather compete for those people.
  - sbacic 5y ago
    This is all contingent on said leetcode riddles actually helping with candidate selection. If they aren't, you've just limited your pool of potential candidates for no reason whatsoever.
    - seanp2k2 5y ago
      https://github.com/poteto/hiring-without-whiteboards https://github.com/poteto/hiring-without-whiteboards
  - jjav 5y ago
    > If I am running the company, I would MUCH rather operate short-handed than "fix" my recruiting problems by lowering the bar. Nobody is talking about lowering the bar, but of putting the bar in the right zipcode. An extremely high bar of leetcode testing is irrelevant to the actual job, so that's an irrelevant bar.
    - mcguire 5y ago
      And yet, those companies seem to be more successful than others with "more relevant" bars.
      - Retric 5y ago
        Leet code interviews only showed up late in these companies trajectories. Internally it’s no longer about what the company wants so much as what people inside the company want which presents huge conflicts of interest. Google for example has done significant research and found such practices wasteful, but internal culture is what it is. In practical terms the FAANG companies internal processes are horrible and they can’t seem to innovate at all, but as long as they continue to print money there is zero reason to risk change.
- dmoy 5y ago
  The riddle-like questions are trash, I agree. Any question that requires you to already know a slightly more arcane data structure or algorithm, or involves just regurgitating some algorithmic trick, or similar, really just gives you signal of "did this person already know the question beforehand?", and nothing else. Not asking those types of questions, and instead asking a more straightforward programming question with relatively simple data structures, and some tradeoffs to be made, plus maybe some follow up optimization questions (but like in general systems terms, not code), tend to work better. But it's still whiteboard coding for the most part. While I do detest leetcode interviews, I've yet to see another interview system that works for extremely large companies, that covers all the bases: * won't result in you hiring someone who can't write a for-loop in their language of choice (I know fizz-buzz is a meme at this point, but I have interviewed people who failed a question easier than fizz-buzz before - not even on a whiteboard. People who genuinely cannot program at all will apply for programming jobs, and some of them will get past the recruiters.) * is reasonably based on skills * is not hyper selecting for "people extremely similar to people who already work here" (not "what school did you go to?" or "who do you know who already works here?" type stuff) * can be made mostly uniform across the company (though, even leetcode-like interviews are hard to make uniform, it's easier than a lot of other methods) * isn't trivially cheated (100% remote tests, where someone else just does it for you) * won't be outright rejected by people who already have jobs (internships, though this does work well for new grads) * won't be outright rejected by people who have less free time outside of the job (e.g. people with families, small kids won't go for "work for a week" type take-home assignments which say "couple hours" but you're competing against people who will dump 40+ hours into it in a single week) A lot of other systems work really well if you're willing to slash your candidate pool to a smaller percentage, but break down once you start trying to fairly get at more candidates. Something that works well for a 1000 person company won't work if you have >50,000 developers. Something that works well for a company explicitly willing to exclude parents won't work well for a super large company. Also, a lot of companies do change up the application process for people who are just out of school, in that they allow internships/etc, with a different route through than just interviews. (Though you usually have to go through an interview to get the internship, they tend to be way easier - having sat on such an intern hiring committee before, you get very few interviewers asking riddle/trick questions) Time-wise, I can generally coach someone to pass a big tech interview loop in 1-3 months, not a year.
- tick_tock_tick 5y ago
  Google has gone on record that the best way to hire would be IQ tests but they don't because of the blowback it would generate. Leetcode is just a shitty proxy for IQ testing.
  - imbnwa 5y ago
    Link to record?
  - sidlls 5y ago
    "Shitty proxy" as in "little to no demonstrable correlation whatsoever." Willingness to grind/memorize is about as poor a measure of intellectual ability as one can get. It does show some level of interest/diligence--but perhaps not the best kind.
  - SketchySeaBeast 5y ago
    That seems to imply that they are willing to train you in whatever you need give you prove you have an aptitude. That too isn't the industry standard as near as I can tell, instead it's becoming just enough of an expert to barely get whatever it is you need done without an understanding of best practice before moving onto the next item.
  - mixmastamyk 5y ago
    Do do do do… What are the issues with IQ tests?
    - alasdair_ 5y ago
      > What are the issues with IQ tests? They can have cultural or social biases unless very carefully calibrated. Ideally you should be able to take anyone who speaks a reasonable level of English, give them test and have them get a score that matches their intelligence percentile. In practice, the tests are built for a particular socioeconomic class and type of person. As a trivial example, imagine the test uses the word "yacht" and you give the test to someone who has never culturally been exposed to that word - that person will, on average, score a bit lower than someone who recognizes the word immediately. Or imagine you give a test using Imperial units to a European who exclusively grew up using Metric (or vice versa). Knowing that three feet are in a yard isn't an intelligence test, it's a test of cultural familiarity.
- theferret 5y ago
  They're not hurting for devs, they're hurting for engineers.
- hinkley 5y ago
  Is it a talent problem, a gatekeeping issue, or are good people just not interested unless the money is insane.

> FAANG are struggling to fill demand in hiring and are offering increasingly high salaries This has been the case, according to my colleagues who are older, since at least 2003 ;-)

vmception 5y ago

> Recruiters are charging 22-25% for placements, and having difficulties filling them. This has been the same as well, to my knowledge, since 2010

rubicon33 5y ago

And yet, they still persist on presenting leetcode riddles that have very little relevance to the job being interviewed for. If they were hurting that much for devs, you'd think they might lower the bar a little by not requiring a years study of leetcode.

kamarg 5y ago
It's more expensive to make a bad hire than to not hire someone and move a bit slower.
- skeeter2020 5y ago
  while this is true, maybe they should work on how they assess candidates instead of continually complaining about how there are none.
- kkdaemas 5y ago
  So... they double-down on a poor proxy of effectiveness on the job?
- lanstin 5y ago
  Leet code is neither sufficient or necessary to writing brilliant software. Your statement is true but Google, being mostly people cut from the same cloth, has group think about these issues. On the other hand, that monolithic culture is probably what has limited their success and kept them from being a more harmful monopoly.
- mech422 5y ago
  google admitted years ago, it didn't help.. https://www.theatlantic.com/business/archive/2013/06/google-finally-admits-that-its-infamous-brainteasers-were-completely-useless-for-hiring/277053/ https://www.theatlantic.com/business/archive/2013/06/google-...
  - fshbbdssbbgdd 5y ago
    Google found that riddles didn’t help, so they stopped asking them. They didn’t find the same for coding questions.
    - mech422 5y ago
      Yet :-P I wouldn't be surprised to see a similar article about leet code...
- marcosdumay 5y ago
  There are two problems with this phrase, one is that it shouldn't. Really, if somebody a bit worse goes way into the negative value for you and you are a large company, you have some procedural problem that should be fixed. Long-time employees also have bad moments, and you should be able to survive those. The second problem is that the onus is on you on making sure those bad interviews decrease the odds if hiring bad people, instead of increasing them or being irrelevant. Without that evidence, this is a non-argument.
- smnplk 5y ago
  Why not train a bad hire ?
  - dagw 5y ago
    Training a 'bad' hire requires taking 'good' employee out of circulation for a while and, depending on what way they are bad, might not lead to anything anyway.
    - dagw 5y ago
      I want to add to this that in many cases getting many "bad hires" can also be symptom of bad management or bad workplace environment.
xyzelement 5y ago
If I am running the company, I would MUCH rather operate short-handed than "fix" my recruiting problems by lowering the bar. You don't have to agree with the bar the companies are using, but that is the bar they have converged on, and one that plenty of people are capable of passing. Would much rather compete for those people.
- sbacic 5y ago
  This is all contingent on said leetcode riddles actually helping with candidate selection. If they aren't, you've just limited your pool of potential candidates for no reason whatsoever.
  - seanp2k2 5y ago
    https://github.com/poteto/hiring-without-whiteboards https://github.com/poteto/hiring-without-whiteboards
- jjav 5y ago
  > If I am running the company, I would MUCH rather operate short-handed than "fix" my recruiting problems by lowering the bar. Nobody is talking about lowering the bar, but of putting the bar in the right zipcode. An extremely high bar of leetcode testing is irrelevant to the actual job, so that's an irrelevant bar.
  - mcguire 5y ago
    And yet, those companies seem to be more successful than others with "more relevant" bars.
    - Retric 5y ago
      Leet code interviews only showed up late in these companies trajectories. Internally it’s no longer about what the company wants so much as what people inside the company want which presents huge conflicts of interest. Google for example has done significant research and found such practices wasteful, but internal culture is what it is. In practical terms the FAANG companies internal processes are horrible and they can’t seem to innovate at all, but as long as they continue to print money there is zero reason to risk change.
dmoy 5y ago
The riddle-like questions are trash, I agree. Any question that requires you to already know a slightly more arcane data structure or algorithm, or involves just regurgitating some algorithmic trick, or similar, really just gives you signal of "did this person already know the question beforehand?", and nothing else. Not asking those types of questions, and instead asking a more straightforward programming question with relatively simple data structures, and some tradeoffs to be made, plus maybe some follow up optimization questions (but like in general systems terms, not code), tend to work better. But it's still whiteboard coding for the most part. While I do detest leetcode interviews, I've yet to see another interview system that works for extremely large companies, that covers all the bases: * won't result in you hiring someone who can't write a for-loop in their language of choice (I know fizz-buzz is a meme at this point, but I have interviewed people who failed a question easier than fizz-buzz before - not even on a whiteboard. People who genuinely cannot program at all will apply for programming jobs, and some of them will get past the recruiters.) * is reasonably based on skills * is not hyper selecting for "people extremely similar to people who already work here" (not "what school did you go to?" or "who do you know who already works here?" type stuff) * can be made mostly uniform across the company (though, even leetcode-like interviews are hard to make uniform, it's easier than a lot of other methods) * isn't trivially cheated (100% remote tests, where someone else just does it for you) * won't be outright rejected by people who already have jobs (internships, though this does work well for new grads) * won't be outright rejected by people who have less free time outside of the job (e.g. people with families, small kids won't go for "work for a week" type take-home assignments which say "couple hours" but you're competing against people who will dump 40+ hours into it in a single week) A lot of other systems work really well if you're willing to slash your candidate pool to a smaller percentage, but break down once you start trying to fairly get at more candidates. Something that works well for a 1000 person company won't work if you have >50,000 developers. Something that works well for a company explicitly willing to exclude parents won't work well for a super large company. Also, a lot of companies do change up the application process for people who are just out of school, in that they allow internships/etc, with a different route through than just interviews. (Though you usually have to go through an interview to get the internship, they tend to be way easier - having sat on such an intern hiring committee before, you get very few interviewers asking riddle/trick questions) Time-wise, I can generally coach someone to pass a big tech interview loop in 1-3 months, not a year.
tick_tock_tick 5y ago
Google has gone on record that the best way to hire would be IQ tests but they don't because of the blowback it would generate. Leetcode is just a shitty proxy for IQ testing.
- imbnwa 5y ago
  Link to record?
- sidlls 5y ago
  "Shitty proxy" as in "little to no demonstrable correlation whatsoever." Willingness to grind/memorize is about as poor a measure of intellectual ability as one can get. It does show some level of interest/diligence--but perhaps not the best kind.
- SketchySeaBeast 5y ago
  That seems to imply that they are willing to train you in whatever you need give you prove you have an aptitude. That too isn't the industry standard as near as I can tell, instead it's becoming just enough of an expert to barely get whatever it is you need done without an understanding of best practice before moving onto the next item.
- mixmastamyk 5y ago
  Do do do do… What are the issues with IQ tests?
  - alasdair_ 5y ago
    > What are the issues with IQ tests? They can have cultural or social biases unless very carefully calibrated. Ideally you should be able to take anyone who speaks a reasonable level of English, give them test and have them get a score that matches their intelligence percentile. In practice, the tests are built for a particular socioeconomic class and type of person. As a trivial example, imagine the test uses the word "yacht" and you give the test to someone who has never culturally been exposed to that word - that person will, on average, score a bit lower than someone who recognizes the word immediately. Or imagine you give a test using Imperial units to a European who exclusively grew up using Metric (or vice versa). Knowing that three feet are in a yard isn't an intelligence test, it's a test of cultural familiarity.
theferret 5y ago
They're not hurting for devs, they're hurting for engineers.
hinkley 5y ago
Is it a talent problem, a gatekeeping issue, or are good people just not interested unless the money is insane.