#Robinhood Karat
1 messages · Page 1 of 1 (latest)
20 mins of "not really systems design but kinda systems design" questions
3 rounds of systems questions, if u finish early you dont get bonus time for code
30 mins of coding, if you finish problem 1 early enough u start problem 2 but i dont think you're supposed to finish problem 2
coding wasnt leetcode it was more of a complex processing problem
Thanks for getting back! What do you mean by complex processing problem? @raw nest
the algorithm isnt something crazy like "oh i figured out the math and i can reduce to O(n)"
its more like the data they give you vs the data you want to generate from it is rough
ill be vague but e.g. question is an unordered list of events, and you have to process those events and determine [some facts about patterns in the events]
Oooh I see, so wording is closer to a real life problem than actual LC
I think it's fair to say that there's a lot of lc problems that are based on that type of event data
like scheduling problems
somewhere along those lines
for ref I got some dp problem so I am definitely envious
are the system design q's like building a system from scratch and describing architecture and all that? or are they like mini questions asking about tradeoffs and stuff
more on the trivia side
oh thats good
but less trivia and more examples/scenario-based I guess
not complicated, it takes like 30 sec to explain each one roughly
should give u an idea of how it compares against a 1-hour sys design interview lol
word lol
that's the intern karat
ah makes sense
would you say if you covered the system topics mentioned in their karat pdf youd be good?
the karat is distributed systems, not system design
the final would be sys design
yeah cover whatever's in the pdf
it's pretty accurate
word
thanks homie
i have mine tomorrow lol
i've taken a distributed course at my uni so hopefully that can carry
coding question second half seemed very "does it work" rather than "is it ideal code" btw
i wrote smth with like, factorial runtime for my part 2
and passed
yeah ive heard others say they only care about completeness
whats the analysis on
you iterate through a list of length n
adding entries to a list
and every time you add an entry to a list
you permute it, where m = current list size, listing every m choose 1, m choose 2... m choose m variation
i wanna say its overall factorial
idk I just feel like a lot of problems have a neat, clean solution
and brute force or raw simulation gets messy
yeah no i agree
whether it's correctness or time/space complexity analysis
i have no idea what my runtime is
perfectly fuckin correct tho
👌
thank god they didnt ask what the runtime was of it
interviewer probably didnt know either
are either of yall still in the process for rh?
got offer last week
oh shii congrats!!
i have onsite in the next week 🤞
no man on earth could decipher the runtime of my code
it might be 2^n not n! actually
fuck idek