EPTThe Big Programming Thread - Page 993
| Forum Index > General Forum |
Thread Rules 1. This is not a "do my homework for me" thread. If you have specific questions, ask, but don't post an assignment or homework problem and expect an exact solution. 2. No recruiting for your cockamamie projects (you won't replace facebook with 3 dudes you found on the internet and $20) 3. If you can't articulate why a language is bad, don't start slinging shit about it. Just remember that nothing is worse than making CSS IE6 compatible. 4. Use [code] tags to format code blocks. | ||
|
Deleted User 3420
24492 Posts
| ||
|
enigmaticcam
United States280 Posts
| ||
|
HornyHerring
Papua New Guinea1059 Posts
On February 01 2019 05:07 enigmaticcam wrote: We hired a SQL DBA once. On paper and interview, he seemed to be very knowledgeable. But after he was hired, when he actually sat down to write a sql query, it was painful to watch. He took a long time to construct even a basic select statement, and it didn't help that his typing was all hunt and peck. He didn't last long, to say the least. That speaks volumes of how good your recruiting process is. Can't imagine hiring someone for a technical position without actually testing their skills in a real scenario. | ||
|
Excludos
Norway8196 Posts
On February 01 2019 06:32 HornyHerring wrote: That speaks volumes of how good your recruiting process is. Can't imagine hiring someone for a technical position without actually testing their skills in a real scenario. Live testing is absolutely worthless beyond finding out if they have basic knowledge (Which, granted, in this scenario that might have worked). For anything beyond a beginner level programmer, tests is a waste of time. You won't learn how good someone is with a 10 minute challenge. Hell, even a 1 hour challenge isn't useful. A much better indicator is looking at someones github history (or any other version control site). Even if it's just a bunch of stuff cobbled together for the interview (which is what I tend to do), it gives a pretty good insight into how structured someone is, how they solve difficult problems, how clean their code is, and all that good jazz. And it lets people show off what they know in their own time. Programming isn't a speedrunning challenge. It doesn't matter if you manage to write fizzbuzz in 10 minutes if you're the type of person who creates large tech debt over time. As a personal anecdote: I fucking suck at these live programming challenges. I spend a lot of time setting up classes, structures and functions for future proofing, and it's a habit I have difficulty throwing away. Yes, of course I can write fizzbuzz in matter of seconds, but I have previously been given 1 hour challenges to "show what I got" for interviews, and failed miserably. Only a short while later I was hired by one of the most highly regarded consultant companies in the city, currently working on scientific projects with bleeding edge technologies (VR, MR and AR among other things), finding solutions Microsoft and Unity themselves are trying to get a hold of. edit: A technological interview with someone's already prewritten code as a background for questions is also a very easy way to find how deep someone's understanding of something is, why they've done something in a certain way, or if they've copy pasted it from StackOverflow | ||
|
Blitzkrieg0
United States13132 Posts
Brief introduction. Fizzbuzz. Then you waste 10 minutes of your day instead of an hour for each of the failures. | ||
|
phar
United States1080 Posts
On February 01 2019 06:50 Excludos wrote: Live testing is absolutely worthless beyond finding out if they have basic knowledge (Which, granted, in this scenario that might have worked). For anything beyond a beginner level programmer, tests is a waste of time. You won't learn how good someone is with a 10 minute challenge. Hell, even a 1 hour challenge isn't useful. ... edit: A technological interview with someone's already prewritten code as a background for questions is also a very easy way to find how deep someone's understanding of something is, why they've done something in a certain way, or if they've copy pasted it from StackOverflow One problem with this approach is that you won't get any applicants who currently work at places with restrictive moonlighting policies, or applicants who do things in their free time other than writing code. If your hiring needs are small, that can be fine to just not interview that pool of applicants. But for a sufficiently large company, this approach does not scale. But yea, 1 hour isn't nearly enough. | ||
|
SC-Shield
Bulgaria832 Posts
On February 01 2019 09:17 Blitzkrieg0 wrote: You absolutely want to do some basic problem like fizzbuzz to weed people out. Just be prepared for the soul crushing that follows when a large number of people can't. Brief introduction. Fizzbuzz. Then you waste 10 minutes of your day instead of an hour for each of the failures. Only for beginners though? I solved Fizzbuzz back when I was graduating from university. On January 31 2019 23:38 Silvanel wrote: Since i am now running our teams testlab and one person just left our company i will be participating in a interviews for hiring new people. First time being on that side of interview i am already quite nervous! Tips? As a colleague says, don't interview people to humiliate them. I think people shouldn't be interviewed for a tiny detail they might not know but can find by going to Google in a matter of seconds. Interview them to see how much they know and don't know. It's probably generic advice but interviewees will also judge you.  Example of tiny detail no one should care about in C++ interview, at destruction you shouldn't care if std::thread will make your program terminate if you didn't join thread. Or, if you use std::async and the instance you got from it will block at destructor until thread finishes. You should just know that you should take care of threads before destructor finishes. I had a question about this stuff at my current company. Obviously, I got hired but it's a meaningless question to ask about details when you already know the "don't do that" rule. It's important to know why you shouldn't do something, but details like this could easily be forgotten over the years, so that's why I say it's meaningless to ask such questions. | ||
|
WolfintheSheep
Canada14127 Posts
| ||
|
Excludos
Norway8196 Posts
On February 01 2019 12:20 phar wrote: One problem with this approach is that you won't get any applicants who currently work at places with restrictive moonlighting policies, or applicants who do things in their free time other than writing code. If your hiring needs are small, that can be fine to just not interview that pool of applicants. But for a sufficiently large company, this approach does not scale. But yea, 1 hour isn't nearly enough. I don't think it's a huge problem to have people cobble togheter something for an interview. It doesn't need to be much, I spent less than 8 hours on it for my current job. If someone is unable to do that then I doubt I would want to hire him anyways. That raises all kinds of red flags | ||
|
Excludos
Norway8196 Posts
On February 01 2019 15:28 WolfintheSheep wrote: Fizzbuzz is to weed out the people lying about bare minimum programming skills on their resume. And sadly it still works. Yeah that's about all it's good for, but I would be surprised if the interviewer didn't manage to catch that in the technical interview. Someone who doesn't know how to program wouldn't be able to answer basic questions about programming either. | ||
|
Acrofales
Spain18132 Posts
On February 01 2019 16:43 Excludos wrote: Yeah that's about all it's good for, but I would be surprised if the interviewer didn't manage to catch that in the technical interview. Someone who doesn't know how to program wouldn't be able to answer basic questions about programming either. Point is that you can dump someone in a cubicle for 15 minutes to go and program fizzbuzz and whoever was supposed to weed that out in the interview doesn't waste his time. | ||
|
Blitzkrieg0
United States13132 Posts
On February 01 2019 14:52 SC-Shield wrote: Only for beginners though? I solved Fizzbuzz back when I was graduating from university. The question was about entry level so in that context yes. You wouldn't ask someone with 10 years experience to come in and do FizzBuzz, but you should have something that you weed people with instead. I work in consulting for PLM Software and we have a list of questions we have people to get a feel for what they've done with the software. Recruiters only care about their bottom line and have project managers interview for developer positions so you need to weed people out quickly. | ||
|
Manit0u
Poland17450 Posts
On February 01 2019 23:38 Blitzkrieg0 wrote: The question was about entry level so in that context yes. You wouldn't ask someone with 10 years experience to come in and do FizzBuzz, but you should have something that you weed people with instead. I work in consulting for PLM Software and we have a list of questions we have people to get a feel for what they've done with the software. Recruiters only care about their bottom line and have project managers interview for developer positions so you need to weed people out quickly. It really depends. We once had a girl come to the interview with 8+ years of experience. We were really excited about her but quite early in the interview during the introductory questions it was revealed that she doesn't even know which version of the framework she's been using all those years... You have to be suspicious of everyone nowadays I guess. | ||
|
Excludos
Norway8196 Posts
On February 02 2019 00:04 Manit0u wrote: It really depends. We once had a girl come to the interview with 8+ years of experience. We were really excited about her but quite early in the interview during the introductory questions it was revealed that she doesn't even know which version of the framework she's been using all those years... You have to be suspicious of everyone nowadays I guess. To be fair that's not something you might remember if you don't actively check it. For instance, I don't have a clue which version of C# I'm currently using without checking. Unless there's incredibly important changes where you absolutely need to know what you're working on (Python 2 vs 3 for instance) I don't think that necessarily reflects your quality as a programmer. It's also easy to forget stuff like that when you're nervous. | ||
|
mahrgell
Germany3943 Posts
I got all kinds of questions about my experience in working as part of a team, various kinds of projects, all kinds of tools (like git and stuff) etc. But never anyone went like "Hey you said you know C++/MPI, how would you do that or why is that like that?" or asked me for some code I had written previously. And in the end I got a positive answer in all but one... | ||
|
Excludos
Norway8196 Posts
On February 02 2019 00:31 mahrgell wrote: I'm kinda amazed by this conversation here. I just recently had a number of entry level interviews with various companies and scientific institutes, mostly in high performance computing and 3D modelling software design... And not one interviewer had ever even tried to test my programming knowledge. I got all kinds of questions about my experience in working as part of a team, various kinds of projects, all kinds of tools (like git and stuff) etc. But never anyone went like "Hey you said you know C++/MPI, how would you do that or why is that like that?" or asked me for some code I had written previously. And in the end I got a positive answer in all but one... So you never had a technical interview is what you're saying? That's quite normal unfortunately, but the company does danger itself of hiring the wrong people this way. | ||
|
mahrgell
Germany3943 Posts
On February 02 2019 00:56 Excludos wrote: So you never had a technical interview is what you're saying? That's quite normal unfortunately, but the company does danger itself of hiring the wrong people this way. If a technical interview requires code work, then no, I never had one, and as I said, there were a number of interviews in very different companies. And at least for the company I work for now, I'm fairly sure they don't end with the wrong people. | ||
|
Manit0u
Poland17450 Posts
On February 02 2019 00:20 Excludos wrote: To be fair that's not something you might remember if you don't actively check it. For instance, I don't have a clue which version of C# I'm currently using without checking. Unless there's incredibly important changes where you absolutely need to know what you're working on (Python 2 vs 3 for instance) I don't think that necessarily reflects your quality as a programmer. It's also easy to forget stuff like that when you're nervous. It was for a web dev position in PHP and the framework you were working with (and its version) is actually quite important since quite often there are quite big differences not just between the frameworks but different versions of it. For example, if it's 2018-2019 and you're still working with Zend 1 or 2 chances are that you're not familiar with the current state of the language and good practices that have been introduced in the past 10 years (and yeah, there are still projects using outdated frameworks like that live in production) and would thus need to be treated as entry level position instead of senior for example due to potential lack of knowledge. | ||
|
Propelled
Denmark184 Posts
| ||
|
Manit0u
Poland17450 Posts
https://www.jetbrains.com/resharper/compare/resharper-vs-visual-studio/ | ||
| ||
