3 ms·

Is it still a huge deal to have an encyclopedic knowledge of data structures and algorithms? This guy was applying for jobs doing React/Node, which is what I do

Is it still a huge deal to have an encyclopedic knowledge of data structures and algorithms? This guy was applying for jobs doing React/Node, which is what I do, and two years after school I cant remember all the details of the sorting algorithms. Would it be so bad to tell the interviewer that if I do anything other than call browser APIs to sort data I am failing at my job and wasting time?

brianwawok 10y ago
You can go that way. I find it very unlikely the person interviewing you would "see the errors in his ways" and go "ok you are right, dumb question". It is a human flaw that when we interview, we project. If we are really good at X, we want to deep dive on that topic. Perhaps it is important for the job. Perhaps we are just trying to show off to a candidate. Who knows. I know I am a fullstack guy, but light on React. If the interview before me was a webdev and asked you a bunch of React questions, I could not add anything of value there. So I would poke around algos and linux and testing and CD and all of the other things I know about, to see where you fall. Not really to show off how leet my algos knowledge is... but I personally would rather hire a guy who knows big O, all things else equal. I do doubt most jobs are so specialized you only need 1-2 frameworks, and nothing else. Knowing what memory is isn't a terrible thing to understand for any programming job...
jaggederest 10y ago
FWIW I've been doing this for over a decade and the number of times I've actually had to care about sort algorithms, balancing red black trees, or that other nonsense is basically only in interviews and only occasionally. I say as much when people try to pull that stuff on me now. There's a copy of TAOCP on the shelf for a reason, if I need to know a good algorithm, Mr. Knuth's are certainly better than mine.
seangrogg 10y ago
It's still a pretty big deal to know how Big O for time complexity (space complexity less so) and how to write algorithms that optimize for such. That being said, I've never had to write anything much more complex than a Binary Search algorithm for an interview as far as DS&A goes. If I were being interviewed I would have no problem taking advantage of any standard APIs I would expect to have available to me, which includes browser APIs if it were a front-end role. Why waste valuable interview time writing a sorting algorithm instead of just using .sort()?
mahyarm 10y ago
It's not very encyclopedic. There are about a dozen questions and data structures that everyone likes to ask. Some places ask harder ones and you have to deep dive into cracking coding the interview. Google is one of them, they have a bigger algorithmic focus. I'd rather have this vs. the 'where did you go school' bias that finance strongly has.

brianwawok 10y ago

You can go that way. I find it very unlikely the person interviewing you would "see the errors in his ways" and go "ok you are right, dumb question". It is a human flaw that when we interview, we project. If we are really good at X, we want to deep dive on that topic. Perhaps it is important for the job. Perhaps we are just trying to show off to a candidate. Who knows. I know I am a fullstack guy, but light on React. If the interview before me was a webdev and asked you a bunch of React questions, I could not add anything of value there. So I would poke around algos and linux and testing and CD and all of the other things I know about, to see where you fall. Not really to show off how leet my algos knowledge is... but I personally would rather hire a guy who knows big O, all things else equal. I do doubt most jobs are so specialized you only need 1-2 frameworks, and nothing else. Knowing what memory is isn't a terrible thing to understand for any programming job...

jaggederest 10y ago

FWIW I've been doing this for over a decade and the number of times I've actually had to care about sort algorithms, balancing red black trees, or that other nonsense is basically only in interviews and only occasionally. I say as much when people try to pull that stuff on me now. There's a copy of TAOCP on the shelf for a reason, if I need to know a good algorithm, Mr. Knuth's are certainly better than mine.

seangrogg 10y ago

It's still a pretty big deal to know how Big O for time complexity (space complexity less so) and how to write algorithms that optimize for such. That being said, I've never had to write anything much more complex than a Binary Search algorithm for an interview as far as DS&A goes. If I were being interviewed I would have no problem taking advantage of any standard APIs I would expect to have available to me, which includes browser APIs if it were a front-end role. Why waste valuable interview time writing a sorting algorithm instead of just using .sort()?

mahyarm 10y ago

It's not very encyclopedic. There are about a dozen questions and data structures that everyone likes to ask. Some places ask harder ones and you have to deep dive into cracking coding the interview. Google is one of them, they have a bigger algorithmic focus. I'd rather have this vs. the 'where did you go school' bias that finance strongly has.