4 ms·

To be clear I think it’s good to test CS fundamentals even if you don’t use them every day. It’s a good screen for aptitude. I’m just concerned that taken to th

To be clear I think it’s good to test CS fundamentals even if you don’t use them every day. It’s a good screen for aptitude. I’m just concerned that taken to the extreme it mistakes excessive pattern recognition prep for true comprehension and aptitude. Especially if it’s “whoever puts the optimal solution on the board fastest wins.” I do think it would do you some good to read up on core CS and algorithms though. I also did not get a CS degree and learned in the field, but now having studied up on it I feel much more in command of my craft. They do come into play now and then, it’s not just brain teasers. It does take weeks to months to really learn it though so I understand the resistance.

dntrkv 7y ago
A strong CS foundation is hugely valuable and I don't think anyone will argue that but the fact that someone with over a decade of experience can go from a "no hire" to "hire" after a couple months of practice tells me that we aren't testing for the right skills.
- abalone 7y ago
  Being exactly of those people (experienced but would have failed algo screen and had to brush up), I can confirm it would have false negatived me despite years of competent productivity... and yet I also feel I have filled in some significant gaps that I was lucky to avoid previously. I think if you have a really strong “portfolio” of individual work that can be validated as being yours, like side projects, that can be a valid pathway into at least non senior roles. Maybe you’re weak on algos and other team members can help out at key points. But I don’t know a better objective and direct measure of coding aptitude inside of 60 minutes. My issue is not with it in principle, at least for senior roles. It’s taking it to an extreme and misinterpreting the results. It gives this false sense of certainty about aptitude.
hello_moto 7y ago
> I’m just concerned that taken to the extreme it mistakes excessive pattern recognition prep for true comprehension and aptitude. Especially if it’s “whoever puts the optimal solution on the board fastest wins.” I agree with you on this. They do have taken this to the extreme because the questions are getting harder and harder. This is hard tbh because if you're an employer whom used to be able to filter candidates by asking "easy" algorithms questions and woke up one day to only find out that everybody knows the answer, what will you do? "It used to work so we just have to improve it, right?" => make a more challenging questions.
- JudgeWapner 7y ago
  >what will you do? You'll do the same thing most other professional industries do. List the requirements for the job: x to y years experience with technologies a,b and c. Grab a handful of the most polished resumes, interview them for personality, and hire the one you liked best.
  - hello_moto 7y ago
    Previous method works well,why change it? Silicon Valley thrive in tricky algo questions. That's how they hired engineers since long time ago.
- abalone 7y ago
  Improvement doesn't mean making it harder. SAT questions don't get harder every year (ideally). They're just different. It takes effort to come up with something different and novel though. Most places don't. Maybe that's what the industry needs.. its own SAT. Heck man, that would make interviewing a lot better. You take one brutal test over a few hours, get your score, and then apply to multiple companies. You don't have to do dozens of whiteboard code interviews. Job postings specify score ranges. There's different score components (algos vs system design, like math vs verbal). I guess there are startups like triplebyte trying to do that, but they're not as independent as the SAT. They make money off hires.
  - hello_moto 7y ago
    I would argue that we are going toward that direction with slightly different variation/format. Faang and multiple SV companies will ask 3-5 LeetCodes with 1-2 System Design. Some started to incorporate behavior question (like Amazon with 14 LPs)
namelosw 7y ago
But CS is not only about algorithms and data structures. Even schools are missing the point mostly. Things like general PLT and formal methods, namely how to deal with abstractions in a large project, and how to know if your program is correct, are mostly ignored because they're not favored by the market.

dntrkv 7y ago

A strong CS foundation is hugely valuable and I don't think anyone will argue that but the fact that someone with over a decade of experience can go from a "no hire" to "hire" after a couple months of practice tells me that we aren't testing for the right skills.

abalone 7y ago
Being exactly of those people (experienced but would have failed algo screen and had to brush up), I can confirm it would have false negatived me despite years of competent productivity... and yet I also feel I have filled in some significant gaps that I was lucky to avoid previously. I think if you have a really strong “portfolio” of individual work that can be validated as being yours, like side projects, that can be a valid pathway into at least non senior roles. Maybe you’re weak on algos and other team members can help out at key points. But I don’t know a better objective and direct measure of coding aptitude inside of 60 minutes. My issue is not with it in principle, at least for senior roles. It’s taking it to an extreme and misinterpreting the results. It gives this false sense of certainty about aptitude.

hello_moto 7y ago

> I’m just concerned that taken to the extreme it mistakes excessive pattern recognition prep for true comprehension and aptitude. Especially if it’s “whoever puts the optimal solution on the board fastest wins.” I agree with you on this. They do have taken this to the extreme because the questions are getting harder and harder. This is hard tbh because if you're an employer whom used to be able to filter candidates by asking "easy" algorithms questions and woke up one day to only find out that everybody knows the answer, what will you do? "It used to work so we just have to improve it, right?" => make a more challenging questions.

JudgeWapner 7y ago
>what will you do? You'll do the same thing most other professional industries do. List the requirements for the job: x to y years experience with technologies a,b and c. Grab a handful of the most polished resumes, interview them for personality, and hire the one you liked best.
- hello_moto 7y ago
  Previous method works well,why change it? Silicon Valley thrive in tricky algo questions. That's how they hired engineers since long time ago.
abalone 7y ago
Improvement doesn't mean making it harder. SAT questions don't get harder every year (ideally). They're just different. It takes effort to come up with something different and novel though. Most places don't. Maybe that's what the industry needs.. its own SAT. Heck man, that would make interviewing a lot better. You take one brutal test over a few hours, get your score, and then apply to multiple companies. You don't have to do dozens of whiteboard code interviews. Job postings specify score ranges. There's different score components (algos vs system design, like math vs verbal). I guess there are startups like triplebyte trying to do that, but they're not as independent as the SAT. They make money off hires.
- hello_moto 7y ago
  I would argue that we are going toward that direction with slightly different variation/format. Faang and multiple SV companies will ask 3-5 LeetCodes with 1-2 System Design. Some started to incorporate behavior question (like Amazon with 14 LPs)

namelosw 7y ago

But CS is not only about algorithms and data structures. Even schools are missing the point mostly. Things like general PLT and formal methods, namely how to deal with abstractions in a large project, and how to know if your program is correct, are mostly ignored because they're not favored by the market.