|
Felime posted:I'm a dummy and posted this first in the working in IT thread. A bunch of questions that may or may not have been answered. If there's resources I missed in the OP then feel free to tell me to gently caress right off. I have a degree in Japanese literature and I have a job writing code. Getting in is the hard part, I suppose.
|
#
?
Dec 17, 2016 00:04
|
|
|
|
| # ? Jun 4, 2024 04:04 |
|
|
Here is a job posting that I had to respond to the dumbass recruiter with just laughs. The title alone is funny.quote:Senior JavaScript Developer Lead
|
|
#
?
Dec 17, 2016 00:40
|
|
|
A certification for JS indeed, strewth.
|
|
#
?
Dec 17, 2016 00:42
|
|
|
The easiest way to get a programming job is to make something good, post about it online, put the code on GitHub, point it and say "I made that." There's a shortage of programmers worldwide. You might have to move on your own dime if you live too far away from jobs but even then it's the same advice; make something good and then let people know you can make good things for them too. If you can provably write code in at least one popular language you can find a job somewhere. It won't happen overnight if you're a noob but really the #1 thing you should do is prove to somebody that you can write good code. They can teach you the rest. Fun fact: most professional programmers don't have a CS degree. Like 10% of programmers never even went to college. At all. For anything.
|
|
#
?
Dec 17, 2016 01:26
|
|
|
Yeah, I dropped out of med school and took a job as a pizza guy to make ends meet. After a decade of odd jobs things picked up (I got add meds) and now, another 12 years later, I am making bank as a test automation engineer. A job so unloved it could be in the IT version of Mike Row's Dirty Job, right between first line helpdesk support and multi-functional repairman. Which is why I am in this niche, not a lot of competition.
|
|
#
?
Dec 17, 2016 19:41
|
|
|
Keetron posted:Yeah, I dropped out of med school and took a job as a pizza guy to make ends meet. After a decade of odd jobs things picked up (I got add meds) and now, another 12 years later, I am making bank as a test automation engineer. A job so unloved it could be in the IT version of Mike Row's Dirty Job, right between first line helpdesk support and multi-functional repairman. Which is why I am in this niche, not a lot of competition. I get that testing is tedious and boring but it really baffles me how much hate it gets. Testing is so, so incredibly important. We run automated tests at work ourselves and the code absolutely does not go into production if it fails. It's incredibly useful.
|
|
#
?
Dec 17, 2016 19:44
|
|
|
Testing is a different skillset than feature development and it's rare for someone to be good at both. Thus most developers aren't very good at testing and don't like to do it, thus it gets a bad rap. Somewhat separately, it's obvious that companies have to hire for feature development or stuff doesn't get built, but it's not always as obvious to smaller shops that they need a dedicated testing team too. At the end of the day there's more development positions out there, and young people going through school are told to look for them. Conversely while there's probably fewer testing positions, there's fewer qualified people applying to them, so it is a niche that works out well for those who enjoy doing it.
|
|
#
?
Dec 17, 2016 20:11
|
|
|
I'd like to know what separates good QA from bad QA. I am a weirdo about unit and integration testing, fwiw.
|
|
#
?
Dec 17, 2016 21:18
|
|
|
WINNINGHARD posted:I'd like to know what separates good QA from bad QA. I am a weirdo about unit and integration testing, fwiw. From my understanding a good QA person tests every path through the code, making sure that improper stuff throws the right exceptions and is handled properly. A good QA person documents bugs, test failures, and whatever in detail. A bad QA person only tests the happy paths and says things like "it crashed when I clicked on *thing.*"
|
|
#
?
Dec 17, 2016 21:52
|
|
|
ToxicSlurpee posted:From my understanding a good QA person tests every path through the code, making sure that improper stuff throws the right exceptions and is handled properly. A good QA person documents bugs, test failures, and whatever in detail. A bad QA person only tests the happy paths and says things like "it crashed when I clicked on *thing.*" For the most part, yes, but I've also experienced the bad QA person that was way too aggressive and could not conceptualize the idea that too much QA can be bad if it leads to slow and infrequent releases.
|
|
#
?
Dec 17, 2016 23:58
|
|
|
Good QA: - Actually tests all the actual acceptance criteria for a feature - Documents bugs in detailed terms of steps taken to reproduce, expected outcome, and actual outcome - Attaches detailed screenshots, error logs, etc to defect logs whenever possible. - Knows enough to test common fault cases ("I double-clicked 'Pay' and it charged twice", "I clicked a button with no network access and it crashed," "I was able to order 'BUTTS' quantity of items when it should be a numeric field," etc) - Has knowledge about the platform they're testing on, to guide them in things to test ("I disabled the Location permission for this Android app in the settings, and then it crashed on launch") Bad QA: - Doesn't actually test against the acceptance criteria, or imposes their own criteria that isn't found anywhere in the original requirements - Logs bug tickets with no detail as to what went wrong, how to reproduce, etc. - Chucks poo poo over the wall and doesn't communicate with developers tasked to fix bugs
|
|
#
?
Dec 18, 2016 08:35
|
|
|
ToxicSlurpee posted:I get that testing is tedious and boring but it really baffles me how much hate it gets. Testing is so, so incredibly important. We run automated tests at work ourselves and the code absolutely does not go into production if it fails. ToxicSlurpee posted:From my understanding a good QA person tests every path through the code, making sure that improper stuff throws the right exceptions and is handled properly. A good QA person documents bugs, test failures, and whatever in detail. A bad QA person only tests the happy paths and says things like "it crashed when I clicked on *thing.*" All in all, a niche, and a boring one in my opinion, because you're not creating anything new, just working with other people's code and doing nothing but dealing with problems. Kind of like help desk, hehe. That's my reasoning why developers hate testing. The only test dev related job I'd want to go into was writing actual unittesting frameworks and test automation tools / scripts, that could be as fun as regular development, because I love frameworks and tools. 
|
|
#
?
Dec 18, 2016 15:46
|
|
|
The best QAs I've worked with have had a wicked strong conceptual model of the system. I remember one time a coworker and I were working on two different features in the same area of the application, and our QA thought of a way to break it before we even merged our branches.
|
|
#
?
Dec 18, 2016 17:26
|
|
|
Keetron posted:Yeah, I dropped out of med school and took a job as a pizza guy to make ends meet. After a decade of odd jobs things picked up (I got add meds) and now, another 12 years later, I am making bank as a test automation engineer. A job so unloved it could be in the IT version of Mike Row's Dirty Job, right between first line helpdesk support and multi-functional repairman. Which is why I am in this niche, not a lot of competition. We had a really good QA guy at my last job who would probably qualify as an "automation engineer", he wrote and managed all the automated tests. He did an amazing job and was well-respected. At my current job, we have an entire QA department that spends all their time drinking on the job and there are constant fights between QA and development. Good QA guys are worth their weight in gold, tbh.
|
|
#
?
Dec 18, 2016 20:32
|
|
|
Thanks for the advice everyone, I'll keep it in mind and try to get back with any results (though probably more questions when I inevitably screw something up).
|
|
#
?
Dec 19, 2016 16:06
|
|
|
Fellatio del Toro posted:I interviewed at one company that said I'd have to get some sort of Oracle Java Programmer certification after starting. Presumably part of their government contract? If Oracle structures their certification program like Microsoft, the company gets a bunch of free licenses if they have enough certified developers.
|
|
#
?
Dec 19, 2016 16:26
|
|
|
Pilsner posted:Automated testing with code requires a lot more technical experience, and I also think it's only at larger companies that care a lot about their development, that they even have dedicated test automation devs hired. Even then, the ratio of devs to test devs will probably be 10 to 1. Furthermore, in many places the devs will write the unit tests themselves and thus don't need dedicated people to do it. I think a good ratio of devs to test devs is around 2:1 (and this does sometimes happen in the wild), and good test devs don't write unit tests unless they are writing test framework code (unit tests are primarily for helping developers write code and for future refactoring, not to assert system behavior). Good test devs focus on things like automated behavioral tests, performance tests, edge case tests, security tests, and chaos monkey testing. A dedicated person writing unit tests makes very little sense (as well as being one of the worst jobs I can imagine).
|
|
#
?
Dec 19, 2016 20:36
|
|
|
Re test chat. I've been looking at maybe trying out some test driven development lately. Do any of you guys have any opinions or insight about it? Note I'm pretty new at programming, and I'm just about to lay the final strokes on babby's first CRUD webapp. 
-Anders fucked around with this message at 21:20 on Dec 19, 2016 |
|
#
?
Dec 19, 2016 21:17
|
|
|
To me, the main value in TDD is laying out a testable design. That's usually one that is built around functions without side effects, where state changes are pushed as far to the edge of the call graph as possible. In a CRUD application, this might mean building near-fully tested code for one page. That'll establish a pattern you can follow to build testable code for the rest of them. Even if it's not actually under test, the code built on that pattern will be of higher quality. It'll also put you in a position to write regression tests that confirm bug reports down the road. Having formal confirmation of a bug in your code is really useful, and also helps you prevent the re-introduction of the problem down the road.
|
|
#
?
Dec 19, 2016 21:34
|
|
|
-Anders posted:Re test chat. I've been looking at maybe trying out some test driven development lately. Do any of you guys have any opinions or insight about it? Making yourself write code that's easy to test basically makes you do object-oriented and engineering/architecture stuff properly. Spaghetti code is just awful, awful, awful to write automated tests for but if you encapsulate everything properly it makes testing easier. Read about object-oriented programming and software engineering if you haven't yet. Also do some planning ahead of time; read about things like service layers and whatever. Your babby's first program is going to be garbage. You'll probably write spaghetti code and not even realize it. You know what? That's totally OK. Crawl before you walk; walk before you run. Write your lovely, awful garbage application then ask yourself "in what was is this lovely and how can I make my next program less lovely?" Embrace failure. You'll learn that way.
|
|
#
?
Dec 20, 2016 01:22
|
|
|
ToxicSlurpee posted:Making yourself write code that's easy to test basically makes you do object-oriented and engineering/architecture stuff properly. Spaghetti code is just awful, awful, awful to write automated tests for but if you encapsulate everything properly it makes testing easier. Read about object-oriented programming and software engineering if you haven't yet. Also do some planning ahead of time; read about things like service layers and whatever. TDD has a place, but as you say, "babby's first program is going to be garbage". Even if they're doing TDD. No substitute for writing lots of code, reviewing lots of code, and thinking about the kind of code that makes it easier to accomplish things and the kind of code that doesn't. There's room for different styles out there. Personally, I use TDD in very limited circumstances, mostly when exploring a problem I don't have an immediate solution in mind for. That is, I prefer a top-down, BDD style of development rather than TDD bottom-up, because bottom-up is extremely painful if you already have the big picture in mind.
|
|
#
?
Dec 20, 2016 15:27
|
|
|
I see so often in this thread there's such high demand for programmers who have real projects and a GitHub account with tons of stuff in it etc. My wife has sent out close to 400 applications and interviewed at some big places (Google, Khan Academy, Coursera) but keeps getting rejected after the final on-site due to 'lack of experience' even after totally crushing the interviews. What gives? She's getting a little dejected and not sure where to turn. Her algorithms/data structures are solid and she's adding more and more cool side projects to her portfolio all the time but still can't land a job. It's her first engineering job after a career change from teaching. Advice?
|
|
#
?
Dec 20, 2016 22:07
|
|
|
Sounds like a cop out excuse. Is she applying to junior positions? If she crushes interviews, I would imagine at least the bigger companies (e.g. Google) would have no problem coaching her onto their teams.
|
|
#
?
Dec 20, 2016 23:34
|
|
|
Probably just sexism if everything else is going well
|
|
#
?
Dec 21, 2016 01:21
|
|
|
She's been applying to all sorts of positions. We don't know what's going on and it's getting really frustrating (and nobody's giving any actionable feedback). She's a better programmer than me and I've already gotten plenty of offers despite only doing decently well in on-sites. Does she just need to keep at it? It's been three months now of full-time job search and studying.
|
|
#
?
Dec 21, 2016 01:54
|
|
|
Cant speak for the rest, but I don't know how you'd get rejected from Google for lack of experience when that isn't one of the criteria. If anything it sounds like a nicer way to say that she didn't do well interviewing.
|
|
#
?
Dec 21, 2016 02:23
|
|
|
asur posted:Cant speak for the rest, but I don't know how you'd get rejected from Google for lack of experience when that isn't one of the criteria. If anything it sounds like a nicer way to say that she didn't do well interviewing. Well, the question is how far did she get at google. If she was rejected at the hiring committee stage then who knows what the real reason is. If she was rejected at the phone screen stage, then yes its about the interview.
|
|
#
?
Dec 21, 2016 02:26
|
|
|
Google's interview process is a mysterious, bizarre thing that nobody really understands. Partly because literally everybody applies to Google so if they want ten people they probably had over a thousand qualified applicants that got most/all of the way through.
|
|
#
?
Dec 21, 2016 02:45
|
|
|
dantheman650 posted:She's been applying to all sorts of positions. We don't know what's going on and it's getting really frustrating (and nobody's giving any actionable feedback). She's a better programmer than me and I've already gotten plenty of offers despite only doing decently well in on-sites. I graduated in May and it took me 6 months to get a job. I'd say she's doing okay and just needs to keep looking. If I had any advice after being on the job hunt for 6 months, I'd say you should only apply directly to companies. Recruiters are worthless. I found a job that pays well on Craigslist. I dunno if she's going through recruiters, but it's worth mentioning. All they did was help me waste 5 months of my life. Grump posted:Sorry. I don't think I was clear in my last post. My problem is most of the support tickets are for issues where I need access to the backend, which I don't have. So I'm pretty much S.O.L. Pretty much all I have access to do is look at the support tickets and think "hmmm....I wonder why this issue is occurring." I only saw one on the list that was a visible front-end issue, and even for that I don't have permission to reply to the client or edit the source code. I just put the CSS code in a txt file and emailed my manager the file, but he didn't do anything with it. I checked at the end of the day and the ticket was still open. Wanted to give an update on this too. This job is starting off slow, but I was overreacting. My department heads aren't the best at finding me work, but we've discussed and they think it's better to spoonfeed me everything very slowly than throw me into anything challenging. Which I think is dumb, but  . I'm watching lots of Lynda videos and getting a lot better at Javascript and Jquery. And i'm getting paid to do so!
teen phone cutie fucked around with this message at 03:15 on Dec 21, 2016 |
|
#
?
Dec 21, 2016 03:09
|
|
|
Grump posted:I graduated in May and it took me 6 months to get a job. I'd say she's doing okay and just needs to keep looking. I agree on both points here.
|
|
#
?
Dec 21, 2016 03:12
|
|
|
Grump posted:If I had any advice after being on the job hunt for 6 months, I'd say you should only apply directly to companies. Recruiters are worthless. I found a job that pays well on Craigslist. That's not entirely true; I got my current job through a recruiter. It was actually loving weird how it happened. I plopped my resume in the inbox of a company that both hires programmers for themselves and for other people. They're a weird combination of an IT contracting company, recruiting company, and custom software company. Other companies have been going through them when they need warm bodies to fill seats. They weren't much interested in me but one of their clients turned out to be. Apparently the story went that they hired contractors whenever they needed code written but noticed that they had a contractor more of than not and really just needed somebody in house that knew the software. In came me! The best advice for job hunting in tech land is to just vomit resumes everywhere. Any company that hires tech nerds is a target. Worst they can do is say "no" and sending an e-mail is effectively free. The real problem is that a ton of companies have created a chicken and egg scenario. They won't hire anybody without experience but won't give you an opportunity to get said experience. Granted some companies have thoroughly onerous requirements that, sometimes, are literally impossible to fill. In other cases they'd like to have somebody with three years senior experience but if all they get is a dozen new grads their choice is to leave the cube empty or grab a noob and train them. This is especially true in places that offer garbage salaries. As much as people would love to hire good programmers for less than $40,000 a year that...is not likely.
|
|
#
?
Dec 21, 2016 03:15
|
|
|
ToxicSlurpee posted:That's not entirely true; I'm mainly saying that based off where I live. Philly has been unkind to entry level web developers this year, imo
|
|
#
?
Dec 21, 2016 03:18
|
|
|
Grump posted:I'm mainly saying that based off where I live. Philly has been unkind to entry level web developers this year, imo Oh yeah, that makes sense...some areas have crap all for code jobs. I'm in Pittsburgh. What's up, fellow PA goon? 
|
|
#
?
Dec 21, 2016 04:14
|
|
|
Felime posted:Probably also a good plan to shake the rust off. Last thing of substance I coded was a little program to merge some text files for Hearts of Iron 3. I wanted the vanilla resources but borders out of the randomizer mod. Nothing spectacular, but there were enough files I did need to put them in a hash table to pair them in a reasonable amount of time, parse and write files in a form readable in another program, etc... Feels less impressive than it sounds, but I can see how having a few examples of actual problem solving would be great for an interviewer to look at. Sorry but this caught my eye and I had to respond. You shouldn't consider this to be unimpressive. That mindset/process of identifying a problem, working out a solution and using code to implement it is absolutely critical to being a developer and is one of the things that is really hard to teach to someone who doesn't already think that way. Make sure you mention this in any interviews, it's pretty important. In addition, it's a completed project that was done purely for your own amusement/convenience - this counts for a lot as well, it shows a genuine interest.
|
|
#
?
Dec 21, 2016 04:32
|
|
|
Grump posted:Wanted to give an update on this too. This job is starting off slow, but I was overreacting. My department heads aren't the best at finding me work, but we've discussed and they think it's better to spoonfeed me everything very slowly than throw me into anything challenging. It's really not dumb per se, they have code that they're responsible for that they don't want you loving up. That's fair enough. They probably don't have a lot for you just yet because they can only give you the super basic stuff for now. Just be patient - isn't this your first industry job?
|
|
#
?
Dec 21, 2016 04:49
|
|
|
I'm a fresh physics PhD graduate trying to move into industry for large-scale data analysis/data science as it is very similar to my doctoral research. (I assume this is the most relevant thread for my questions - unless someone can point me elsewhere). I would like to include some sort of "job title" in my resume, under my name. Does anyone know if "Data Analyst" or "Data Scientist" would be too presumptuous? Would "Doctor of Philosophy in Physics" be better? Or something else? I know the point of a resume is essentially to keep someone's eyes on it for as long as possible. Lots of places mention in the requirements "advanced degree in relevant field such as blah, blah, blah, physics, blah" but plenty don't mention it and I'm worried the people hiring just...wouldn't be aware that it's a relevant field? Beyond that, any other advice for transitioning to industry would be greatly appreciated.
|
|
#
?
Dec 21, 2016 05:54
|
|
|
FaradayCage posted:I'm a fresh physics PhD graduate trying to move into industry for large-scale data analysis/data science as it is very similar to my doctoral research. (I assume this is the most relevant thread for my questions - unless someone can point me elsewhere). I don't get it, are you asking whether you should make up a job title you don't actually have and put it on your resume?
|
|
#
?
Dec 21, 2016 06:08
|
|
|
The Wizard of Poz posted:I don't get it, are you asking whether you should make up a job title you don't actually have and put it on your resume? More of a "what I (can) do" title than a job title. If it's a job title, then I would put "Self-employed in the industry of job-hunting" which wouldn't be very helpful.
|
|
#
?
Dec 21, 2016 07:26
|
|
|
You don't need a job title under your name, the fact that you're applying for a job explains what you want your next role to be. Just put Graduate Student Researcher (or whatever your school calls it) next to your school's name and your degree. Everyone will understand what you mean. As for transitioning, my advice is: -know some stats (a/b testing, Bayes theorem). Missing the basic stats questions is a way to guarantee failure at the phone screen. -know a programming language that's not Fortran or whatever specialized language is popular in your subfield. Python is a good one to start with. If your only experience coding is with a massive C++ codebase with no version control, forget everything you've learned and start with Python or Java (I'm exaggerating, but only a little). -know some SQL (case statements, joins, etc) -be familiar with basic machine learning algorithms. Recommendation engines, random forests, etc. Ignore neural networks, deep learning, computer vision, and NLP unless you are prepared to spend months on that subject alone or worked on it for your degree. in_cahoots fucked around with this message at 07:45 on Dec 21, 2016 |
|
#
?
Dec 21, 2016 07:37
|
|
|
|
| # ? Jun 4, 2024 04:04 |
|
|
FaradayCage posted:I'm a fresh physics PhD graduate trying to move into industry for large-scale data analysis/data science as it is very similar to my doctoral research. (I assume this is the most relevant thread for my questions - unless someone can point me elsewhere). On my resume I have: Doctoral Researcher of [redacted]. My theory behind it was that someone would look at it and be like "What the gently caress is that" long enough to see the rest of my resume. At my interview my now current boss was really impressed with it, but he turned out to have a PhD in something similar. I think you do need a job title, or a position title, if only because PhD's are still sort of mythical to people, and if you can't quantify it into something people can wrap their heads around they might think you went to school for five years to do voodoo magic. gay for gacha fucked around with this message at 16:53 on Dec 21, 2016 |
|
#
?
Dec 21, 2016 16:37
|
|
