|
the best ideas never win when you have idiots in charge, only the ideas that come from people who are bigger friends with the C suite.
|
#
?
Mar 9, 2019 09:45
|
|
|
|
| # ? May 25, 2024 22:52 |
|
|
Ciaphas posted:i guess the lesson to be taken here is I may or may not be any good at touching computer, but interviewing remains garbage  E: I spent most of my (successful, as it turns out) second interview babbling about how the angle of the sun during rush hour at a curve in the freeway might cause a locally well-known daily traffic jam. I am very much not an engineer and it was very much not a highway engineering role. Interviewing is complete garbage, much like my posts. E2: If H.P. (or other traffic-talking-mans) is reading this thread: is angle of the sun relative to the road surface at high traffic times looked at during road construction? Most people I know don’t seem to like to tear-rear end down the freeway at 70 per with the sun in they eyes? Schadenboner fucked around with this message at 13:00 on Mar 9, 2019 |
|
#
?
Mar 9, 2019 12:25
|
|
|
i liked the onsite i went through. the "can you write computer code" question wasn't a toy problem but didn't include any algorithmic shenanigans, and the "can you algorithm" question didn't require any code (like i started but stumbled on a fencepost and it didn't matter). it really made me realize that tying together "can you think of a solution" with "can you implement a solution" multiplies their respective stresses and i think is a huge contributor to what makes technicals hell.
|
|
#
?
Mar 9, 2019 15:46
|
|
|
qhat posted:at my old org i used to think that the prevalence of tech debt would be a good thing for me since there'd be a lot of opportunity to spearhead some great initiatives, like devops and the like, and then getting recognized as a highly valuable employee. at least that's what i thought until i realized that the tech debt also prevented the good ideas from being taken seriously at all. tech debt is the reification of org priorities it exists as a conscious or unconscious choice of the org and work contra org culture is not going to be valued unless you are hired by a c-level and have their full support (and can deliver quantifiable gains quickly), no ones going to value linting on commit when you should be shipping for revenue
|
|
#
?
Mar 9, 2019 16:00
|
|
|
the poisonous culture that allowed the tech debt to accumulate is also extremely resistant to having someone come in and attack that debt (because it is also an attack on that culture, and specifically the people who created and perpetuate that culture) im about ready to declare a perfectly inverse relationship between the degree a job posting/recruiter/interviewer talks about how they're looking to hire fresh, unconventional, out-of-the-box thinkers and the degree to which they'll be able to propose changes and try doing things differently
|
|
#
?
Mar 9, 2019 16:43
|
|
|
tech debt is almost always a financial issue. nobody wants to spend the money now to improve things for the long term
|
|
#
?
Mar 9, 2019 16:50
|
|
|
Shaggar posted:tech debt is almost always a financial issue. nobody wants to spend the money now to improve things for the long term long term doesn’t exist in publicly traded companies anymore. next quarter revenue or bust!
|
|
#
?
Mar 9, 2019 17:10
|
|
|
it also doesn't exist in privately held companies
|
|
#
?
Mar 9, 2019 17:13
|
|
|
it's one thing to convert a random recursive function into an iterative one, it's completely another to do it for a fairly involved algorithmic problem (BST reversal) if I was ever doing the latter as part of my job I'd definitely use google to help
|
|
#
?
Mar 9, 2019 17:21
|
|
|
g'morning i find this tech debt/code rot discussion interesting, partly because most of my accomplishments the last 10 years were around trying to resolve those via rewrites and the like, so I think about them a lot; and partly because the company i'm interviewing with mentioned similar refit-type work in their pipeline (also they're privately owned, re: the latest discussion)
|
|
#
?
Mar 9, 2019 17:28
|
|
|
I've basically decided to never work for a company again that doesn't have some kind of established process already in place when it comes to integration and deployment. I fixed the issue completely at my old company and only received a token reward, so it definitely wasn't worth it
|
|
#
?
Mar 9, 2019 19:03
|
|
|
so on the topic of "what would make a programming interview worthwhile", I had an onsite yesterday that was pretty interesting and different. I had to bring my own laptop in with a certain set of dependencies preinstalled, and there were questions where they'd give me a repo with a bug in it and had me track it down. This seems way more in line with day-to-day software work, and I'd have been on board with this except that what language I'd like to use and, because it's easy to whiteboard, I say "oh I usually interview in python", so I had to figure out how to use a python debugger right on the spot in front of the interviewer  Still, though, seems less artificial and would be cool if that became the default.
|
|
#
?
Mar 9, 2019 20:09
|
|
|
the big downside there is that you're requiring each candidate to have a laptop and time to janitor it to prepare for the interview which, like, tradeoffs. everything has tradeoffs. so many of them seem to suck.
|
|
#
?
Mar 9, 2019 20:33
|
|
|
raminasi posted:the big downside there is that you're requiring each candidate to have a laptop and time to janitor it to prepare for the interview I had one that also had me do a coding exercise on a laptop, and they offered me the choice of bringing my own or using a pre-set up provided one which seems like an okay option. like sure maybe you have some advantage if you have your own environment set up just how you like but also that’s gonna be the most like you’re gonna use day to day
|
|
#
?
Mar 9, 2019 20:39
|
|
|
I had one that gave me a loving MacBook and I had to stop every two seconds to ask how to do esoteric tasks like "hit F12."
|
|
#
?
Mar 9, 2019 20:42
|
|
|
raminasi posted:the big downside there is that you're requiring each candidate to have a laptop and time to janitor it to prepare for the interview (the bigger problem was that my "ancient" mac doesn't have modern video-out so I couldn't connect it to the conference room's monitor, so I needed an extra 30 seconds to dial into a zoom meeting, but even if I couldn't have done that they would have provided me a laptop anyway  )
|
|
#
?
Mar 9, 2019 21:03
|
|
|
software-engineering-adjacent roles seem to be a lot better with respect to interviews. i interviewed with raytheon for a binary exploitation role at the end of 2017, and we just did some binexp stuff together and it was dope. they just didn't wanna throw enough money at me to make moving to the east coast and doing three-letter contracting worth it so it didn't work out. i also got no-offer from an SF company in 2018, but their interview process was totally fine, it was just that they went with an internal transfer from the company that had just acquired them, and that person had a shitton of securing-big-data-pipeline experience that i didnt have. but i was super positive coming out of the interview, and if they had called me up before i started doing bigtech interviews i almost certainly would've taken an offer from them i had similar results in my most recent crop of interviews: one for google (security engineer), one for amazon (system development engineer aka SRE), and one for a local ~450 person company where i would've basically been The Security Guy and starting their program from the ground up. anyway, the google and AWS interviews were generally reasonable and were obviously connected to the actual work being done. the local one was also totally reasonable, but less detailed technical discussion and more high-level. the google one was the hardest, but there wasn't any "sort a list with X constraints using the One Weird Trick you have to know" bullshit like a lot of the software engineering bigtech jobs require basically i think if you have a specialty that builds off a development background, you don't have to deal with all the algorithm bullshit because they can just ask you about CBC padding oracles or whatever instead, so interviewing is a way more pleasant experience
|
|
#
?
Mar 10, 2019 01:13
|
|
|
I've always wanted to believe that binexp interviews would be super rad, glad to hear that's actually the case 
|
|
#
?
Mar 10, 2019 02:24
|
|
|
Shaggar posted:tech debt is almost always a financial issue. nobody wants to spend the money now to improve things for the long term Chris Roberts account spotted
|
|
#
?
Mar 10, 2019 02:26
|
|
|
CPColin posted:I had one that gave me a loving MacBook and I had to stop every two seconds to ask how to do esoteric tasks like "hit F12." I had this happen to me, and I looked like an idiot being like, Alright, how do I open a web browser and a terminal? Pretty sure it was part of why they decided to pass on me. Culture fit!!
|
|
#
?
Mar 10, 2019 04:09
|
|
|
Had a decent interview last Monday and on COB Friday they responded asking for references and a call with their company's president. This is for starting up a small new department in a very established 100-person company. Not a sure bet but I'm in the running, so that's nice. It's been a dog's age since I've thought about references. Anyone have thoughts on that? So far for #1,#2 I've got my two best friends who are engineers and have known me for a long time. For #3, I've been having a hell of a time tracking down my grad school project mentor. The backup plan is a boss that I had a long time ago who was a pretty big deal but now retired.
|
|
#
?
Mar 10, 2019 16:32
|
|
|
Double post YES!
|
|
#
?
Mar 10, 2019 16:32
|
|
|
Woodstock posted:Had a decent interview last Monday and on COB Friday they responded asking for references and a call with their company's president. This is for starting up a small new department in a very established 100-person company. Not a sure bet but I'm in the running, so that's nice. I think the best references are former bosses you have a good relationship with personally; but I also don't know how much stock anyone puts in references... If you can find 2 or 3 people to say "yeah they're fine" then I think you've checked that box. The fact that you're gonna be talking to the company president probably means you've more or less sealed the deal. My references are a former boss I have a good relationship with, a former head architect I worked with and a long time friend/coworker software engineer.
|
|
#
?
Mar 10, 2019 17:43
|
|
|
cheque_some posted:I had this happen to me, and I looked like an idiot being like, Alright, how do I open a web browser and a terminal? this is the exact reason i took a mac at my current job, so I could learn enough of it to become more OS agnostic. ive used windows or linux stuff forever, but 97% of the company uses macs and the ones who don't always have problems integrating with stuff. thankfully they didn't give me a computer during the interview or i would have made the same mistakes you did. pretty bullshit reason to pass on a candidate
|
|
#
?
Mar 10, 2019 19:24
|
|
|
how do you interview complete newbies, guys fresh out of college with zero work experience? where do I start?
|
|
#
?
Mar 11, 2019 01:32
|
|
|
hackbunny posted:how do you interview complete newbies, guys fresh out of college with zero work experience? where do I start? 0 work experience shouldn't mean 0 programming experience. ask about projects. school projects, side projects, whatever. also this is when those lame-rear end algorithms questions come in actually unironically handy
|
|
#
?
Mar 11, 2019 01:37
|
|
|
bob dobbs is dead posted:0 work experience shouldn't mean 0 programming experience. ask about projects. school projects, side projects, whatever. also this is when those lame-rear end algorithms questions come in actually unironically handy To add to that, having interviewed a million college kids: If they've only done school assignments, almost certainly don't hire them. Even if their side projects are silly trivial things, that's still ok. Find out what they're passionate about or interested in and see if they can apply that to whatever their day-to-day would be. otherwise, here's a handful of stuff I'll usually ask, copy pasta'd from an old text file CS Concepts from School: Data Structures, Algorithms, Big-O run time, recursion How does a Set work? HashMap? What design/coding choices make code easier to test? How do you test performance? Where do you normally see performance issues in the code? How would you go about diagnosing a performance problem in an application? What is an example of a project/code you are proud of? What's the most difficult programming challenge you have experienced? problems you have seen in the real world that you think you could solve The Bank Account Concurrency problem How could simultaneous deposits in a multi-threaded environment cause issues for maintaining an accurate balance What are the different ways you could make this thread safe How does Google Maps work? When I type 123 sesame st into google, how does it know what address I want?
|
|
#
?
Mar 11, 2019 02:37
|
|
|
most new college kids are weakest in testing ime. poke at their testing thoughts and poo poo
|
|
#
?
Mar 11, 2019 02:49
|
|
|
Feisty-Cadaver posted:To add to that, having interviewed a million college kids: If they've only done school assignments, almost certainly don't hire them. Even if their side projects are silly trivial things, that's still ok. Find out what they're passionate about or interested in and see if they can apply that to whatever their day-to-day would be. Ya this is a big one. School projects, unless they are group projects, are very small in value and every rear end in a top hat in their class was forced to do the same thing. If the person went out and learned about stuff on their own, that's a good sign.
|
|
#
?
Mar 11, 2019 02:58
|
|
|
bob dobbs is dead posted:most new college kids are weakest in testing ime. poke at their testing thoughts and poo poo lol I wrote my first test a week ago
|
|
#
?
Mar 11, 2019 03:29
|
|
|
my biggest problem with people straight out of school has always been cowboy coders. absolutely make sure they’re competent but after that i want to know if they’re going to understand that they’re scrubs that don’t know poo poo or if they’re going to cop an attitude where nothing is their fault.
|
|
#
?
Mar 11, 2019 03:37
|
|
|
Rex-Goliath posted:my biggest problem with people straight out of school has always been cowboy coders. absolutely make sure they’re competent but after that i want to know if they’re going to understand that they’re scrubs that don’t know poo poo or if they’re going to cop an attitude where nothing is their fault. My biggest thing, when testing in any language, do you know what (for example, C here) this represents: /* */ I don't do interviews but this would be my top question.
|
|
#
?
Mar 11, 2019 03:46
|
|
|
Feisty-Cadaver posted:To add to that, having interviewed a million college kids: If they've only done school assignments, almost certainly don't hire them. Even if their side projects are silly trivial things, that's still ok. Find out what they're passionate about or interested in and see if they can apply that to whatever their day-to-day would be. if only there was a word in the english language for "an activity that a person with a lot of money demands another person to routinely perform before they will give that person a tiny portion of their money"
|
|
#
?
Mar 11, 2019 05:12
|
|
|
college kids usually aren't quite that jaded yet
|
|
#
?
Mar 11, 2019 05:22
|
|
|
i know, that's why companies love shoving as many of them into the meat grinder as possible
|
|
#
?
Mar 11, 2019 05:25
|
|
|
hackbunny posted:how do you interview complete newbies, guys fresh out of college with zero work experience? where do I start? hire the first one that knows how to use source control and understands why they should never, ever force push
|
|
#
?
Mar 11, 2019 05:28
|
|
|
Stringent posted:hire the first one that knows how to use source control and understands why they should never, ever force push we use force push for squashing commits in the pr or branch before merge because some of the history has already been pushed to the remote maybe the policy should be that the merger of the pr does the squash? not sure here
|
|
#
?
Mar 11, 2019 10:20
|
|
|
lol scm jargon
|
|
#
?
Mar 11, 2019 12:32
|
|
|
dragon enthusiast posted:lol scm jargon
|
|
#
?
Mar 11, 2019 12:39
|
|
|
|
| # ? May 25, 2024 22:52 |
|
|
I need some more figgies in my life imo
|
|
#
?
Mar 11, 2019 13:20
|
|
