Sometimes it's the inspiration to continue working on the new thing until it reaches parity.

But yeah, otherwise, it fails if and only if the exec gets fired which is rare for all of the usual (good and bad) reasons.

maybe not news to most of you, but I found this interesting:

That was really interesting. Thanks for posting.

oh man that interview was hard.

4 separate interviews - one was a quick sort problem where he wanted me to find the k'th largest element, i knew the solution was modified quick sort with a heap implementation but that was about it.

then he wanted me, given two sorted arrays, to find the k'th largest sum between the two arrays. i found the naive solution very quickly but needed a lot of help with the O(nlog(k)) implementation he wanted.

that guy told me i did very well though.

the second interview was coding with the guy I've been interacting with through email this entire time - he gave me a softball mult(int a, int b) question which thank god I've coded a million times and I knocked that one out of the park. he tried to make it more difficult by making me analyze the function, like asking what downside my recursive solution had. I told him it uses a lot of space on the stack, he asked me how to minimize that and I gave the correct answer.

then he went on to an array multiplication problem and wanted to know how i'd optimize it for 4 cores given a 2x2 multiplication. I couldnt remember how to multiple an array but told him we'd need a nested loop running for each calculation in the 2x2 array.

I think those interviews went okay but I felt very lost a lot of the time other than on the softball question. after the coding he was probing me on questions testing my OS knowledge - given a black box program running on an OS, if we realize it's running slower than expected, what are the possible reasons for this and how can we check for them?

i told him 1 signal would be memory utilization - and he asked me why that would slow it down, and i told him once RAM is full then the task scheduler needs to start making decisions about what to switch in and out of memory. he asked what a cause for that could be - and i said perhaps some memory leak in the black box. he liked that answer.

the second signal i gave was cpu utilization, and the third i almost arrived at but it was I/O.

The one ****ing area on my resume that I know a lot about, compilers and programming language analysis they didn't really probe me much on.

the third interview was with the guy who I assume has the final say - he was the co-founder. he was very friendly and broke down what is expected of me day to day and what the environment is like. He asked me what my 2, 5, 10, and 20 year plans were.

He asked me given the two career paths - management and technical knowledge - which one was i likely to follow?

I said that given my work experience (boat captain) i am well versed and comfortable in a leadership role, but given that I am expanding into a new field at my slightly older age, I want to fill in some of the gaps in my technical knowledge, so that would be the path I choose for now.

he was extremely chatty, smiley and friendly despite prefacing the interview that he didn't have a lot of time and work was crazy at the moment. He lingered longer than he had to and answered some of the questions I had about their products. he gave me his business card at the end.

the 4th interview was a ****ing nightmare, it was with their head database guy. Like I said earlier, DB's are my achilles heel. I think I showed enough familiarity with the subject to pass, but I needed a lot of help, especially when we got to transactions. He broke down how their schema worked and how they have managed to save a lot of time and processing power (it's based on flash memory). I gave the correct answer to a few of his questions, but he was an extremely awkward dude with a thick swedish accent so there were a lot of moments of silence I tried to fill as best as I could.

towards the end I think I asked a smart question because he lit up and talked for a while - I asked that if their existing schema is based on flash memory, how do they deal with the read/write overhead that is really hard on SSD's? And he went into a long explanation of their techniques. I asked him a few questions about his background because he seemed like he was really nervous and shy and we chatted for a while.

then the main email guy came back and told me that was all they had, i asked what the next step was and he said they'd let me know in a week.

i could've done a lot better but all in all, i think I can chalk that one up to a passing grade. one positive - i was able to answer all the time/space complexity questions correctly and quickly, which is a strength of mine. There were a lot of those types of questions.

All sounds pretty good to me

Yea dude sounds like you did very well

It did not feel that way - the email guy was giving strong vibes like he didn’t like me. Could just be his personality.

I felt really lost and unprepared. It also went about an hour shorter than was planned.

Noooo, you learned this in algorithms! It's quick select in O(n)! The naive approach is to sort and it be O(nlogn)

I could even picture the exact lecture slide in my head but i couldnt remember

Also what is mult(int a, int b) ? Multiplication of two integers? Without using multiplication?

Yea pretty sure I’ve had that as a test question no less than 3 times. I tried to pretend like i had never seen it before.

I think he was aware of that so he started asking harder questions like what return type should be and i correctly said it needs to be aware of integer overflow etc.

Did you use bit-shifting if what ever number lowest is even?

ie.
4 8 //4 is even so we can bit shift left for a and bit shift right for b
2 16
1 32

stop when a is 1.

no lol it was something like this:

to minimize recursive calls, I said we needed to add a check for a < b and switch the variables.

when he asked me what the run time was i told him it's O(1) because integers have a set size as a head nod to this thread's discussion about my hw problem, lol

but then I told him it's O(b) assuming the integers were in the range of -inf to inf


one other Q i can remember, he asked me to explain B trees to him, i somehow BS'd the answer and he asked why they are beneficial for existing DB systems, i told him because if you are doing a lot of i/o on a hard disk drive it's beneficial if they are physically close to each other on the disk because access times are really slow, he asked how slow, and i said i couldnt remember but i thought it was measured in millseconds. he was happy with that answer

0 Testing, but tail recursion is the answer

My attempt

In before gamingmouse or someone drops a python one liner

Blind app has really been an eye-opener on the salaries stuff. I really need to try and join a bigger company soon after my 1 year. Not getting any stock refreshers is putting me behind and the paper money I'm getting isn't worth much unless the company 3-4x.

I might be totally wrong here, but here's some thoughts that may or may not be helpful/relevant.

It seems like you expect yourself to get everything 100% that you should know in your mind. Meaning, if you have been taught something at some point, researched it on your own, practiced it on a coding site, and are asked about it in an interview, you expect yourself to be able to get it, or consider yourself in the wrong.

It is possible this is interpreted as you thinking you should have every answer and being over-confident. When junior-level people seem over-confident, the oldest play in the interviewing book is to start testing their limits and asking them things they most definitely don't know/ won't get correct. It is sometimes used as a way to disqualify someone (this is the interpretation you seem to have) but also used as a way to see how someone reacts when they don't know/remember something (if they will ask questions/research effectively on the job).

Often times people are expecting you to say you don't know, and then talk it out logically. You were the one there in person, so you know way better, but maybe this is some of what is happening to you and you shouldn't be so hard on yourself.

Oh believe me there was plenty of "I don't know" in the interview, lol.

i will keep that in mind. my therapist tells me I am too humble so I don't think I'm accidentally coming across as arrogant, but maybe that's the case.

overall though looking back the interaction was very friendly and back and forth. it did seem like they were trying to find gaps in my knowledge.

overall it seemed like a really great company and one I'd like to work for if i was extended an offer. the co-founder told me I would learn more in 3 months working there than in the entirety of my undergrad and I believed him.

when I asked him what type of role was he looking to fit me in, he said it would depend on my strong areas - he said they needed all sorts of roles, GUI stuff, DB stuff, and especially compiler type of stuff. Which was funny because that was one area they seemed to specifically avoid when questioning.

Also dont run this with a -b



Sounds like you did very well. Again seems like a harder than average interview tbh. Is this another gaming job? Its pretty funny how much youd destroy an average java/enterprise interview

It’s my embedded systems professor’s company - he asked the class if there were any graduating seniors strong in C/C++ to send him resumes so i did. I had no idea what type of company it was until later. It’s a big data company.

I really wish I had met with my professor today, he really likes me but probably didnt know I was interviewing.

Send him an email letting him know you had the intetview, etc.

he has not connected my name/face - just that i sit in the front row and always answer questions

if i take that route, i may introduce myself on tuesday

What's Blind?

My company moved recently-ish to doing yearly reviews & pay adjustments, and just found out today I'd be getting a raise of ~10% this year. Which is nice and all, but I didn't get much of a raise last year, and if you annualize my pay increases over the time I've been at this company it's gone up about 8% per year, and I'm a senior developer (been doing this 10 years now) not making close to the kinds of money the Google guy is talking about in that video - my base salary is, but lol at the bonuses or massive stock grants that he talks about. (thanks for posting Wolfram) (at a "startup" that's grown from 80 to 200 employees during my time here, so I have options, but they're still not worth much and who knows if they ever will, I'd def be happier with ca$h/liquid stock)

Feels like, as I enter what is probably the prime earning period of my life, I'm not at all maximizing my earning potential. And that's my fault, I haven't been aggressive about feeling out the marketplace, I was at my last job for 6 years and followed a former coworker to my current job without interviewing anywhere else. But given that I'm probably settling down soon and (god forbid) probably going to have to buy a house in this ungodly expensive place, I don't think I can afford to be so complacent anymore.

Feels weird to complain about this when I/we make stupid amounts of money compared to 98% of jobs.

lol I pretty much dont know anything in this post. pretty sure I could google just about all of it though.

do ppl here actually sit down and implement a heap sort at their job?

anyway, sounds like you got the job. if not, then thats just rediculous.

Basically all I do at my job is move numbers from one column to another. There's not much need for any algorithmic design there. Just clear code that is easy to read and maintain and highly likely to work, because if it doesn't people lose money.

Our greatest speed bottleneck is the db, not runtime code.