How much math and what languages is needed to get a code monkey job...

How much math and what languages is needed to get a code monkey job? How much math and what languages is needed to get a well paying programming job?

Attached: 1539824054944.jpg (874x816, 123K)

Other urls found in this thread:

stackoverflow.com/jobs
twitter.com/NSFWRedditVideo

>well paying
fuck off we're full

t. Raj

Trig/Geometry/Statistics/Linear algebra/Graph theory

Study these well and you're on your way to 300K starting

t. Sanjay

Code monkey:
>Javascript/Python/Java/php/Ruby
>None

Well paying:
>Java/Python/C++/C
>Linear Algebra, Graph theory, Statistics, Calculus

math is useless, if you are doing computer science, change to software engineering, it makes you more marketable

t. computer science dropout that later changed to software engineering

You're looking at web dev.

Do the 'Interneting is Hard' guides to CSS, JS, HTML. Use VS Code. Learn a front-end framework after getting comfy with the above. I would suggest React or Vue. At some point learning these, get your head around using npm. Now it's just a matter of using npm packages and copying websites until you're good.

Start with a homepage for yourself with a blog, then something like twitter that has a back-end for storing data.

After you get to that point it branches a bit. Topics like:
- hosting / linux servers (like digital ocean)
- git
- UI frameworks
- databases

etc... you don't have to put a ton of time into most of these things, just understanding the point of them and knowing what documentation pages to read can be enough to start working. If you actually put effort into it you can know enough to host your own websites and sell your labor in a month.

T. Brainlet who couldn't into math who is a shitty code monkey brogrammer hedonist.

Any decent CS or SE degrees should have a lot of overlap with each other. Theory and practical skills are both essential, and learning one but not the other is a complete waste of time.

at least i make 6 figures, good job with your research grants, i'm sure that your math skills are good for something in academia

square root of a piece of pie

t. Shupra

>math and programming
Avoid this path unless you want to compete for a data science job against hard science grads with advanced degrees who couldn't get a job in academia.

Learn databases or webdev.

pretty much this

this is the final redpill to end all bluepills, do db for a comfy corporate job or do webdev which can be associated with nearly everything and earn 100k+ by being just a LITTLE decent.

be able to count
any popular language

Can you explain this to me more?

I always heard that you need a lot of math to do good in programming? Aren't data science jobs the highest paying too so wouldn't a hard science grad have more difficulty getting that job?

>2018
>using math at all in code
>not going pure logic
Dont be one of those java object oriented dinosaurs
Math was killed in the 90s, no one likes static bug-ridden programs

And drop those Java/Object-oriented blinkers they teach in schools
Unless you work for tiny biz everyone wants dynamic, modern, logical code these days
Which is actually easier to learn because logic is natural to learn, while math is brain garbage
Math can't even handle the most static data in the world (age of birth) because it's unable to deal with all the variability involved

Seriously I fired 4 people because they filled their code with math and were unable to "unlearn" school-grade memecoding and learn logic
Most IT heads in any company of medium size and above think like this too
If you want a future abandon math

Most programming doesn't really require math. One major exception would be video game graphics programming.

Ignore most posts in this thread.

Must important thing is you make connections in school.

This 100%

dumb question, should I learn all popular frameworks before looking for work or is it possible to learn while working?

>I always heard that you need a lot of math to do good in programming?
Nope. You just need to start programming and you'll learn the math you need as you go.

>aren't data science jobs the highest paying
Senior engineering positions pay about the same with far more stability across verticals. Data Science is an open ended title which varies from place to place even within the same vertical. In some companies, data science is big data engineering plus some analytics, in other places it's hardcore NLP and in other places it's computer vision.

>hard science grad
>difficulty
The field is so open ended all anyone cares about at this point is the degrees.

It's not a field you want to be in if you have a choice. I'd suggest ML engineering or something else as it's far safer and because the problems have a far narrower scope so you can master the relevant knowledge to have a career not just a job.

What the fuck are you talking about?

>no one wants oo code today
The only people who have embraced FP are newer programming fields like data engineering which are oriented more to data transforms than structure. Everyone else is still using OO because it's the best programming paradigm for their tasks.

>most it heads
Most IT heads know their business. Doing FP because it's the latest niche isn't what anyone with any sense does.

Last time i checked math is fucking hard. Stop being such "i'm above all" retard.

>math is for actual autists
fixed

Not op but my story.
>fuck those web dev faggots ill do cs and be do data analyst.
>see the problems and the math requirements.
>crawl back quietly to learning php.

>How much math and what languages is needed to get a well paying programming job?
A well paying job will require a BSc in Computer Science.
Computer Science is a branch of mathematics (not a branch of computers or anything else), with emphasis on Algebra, Statistics, some Geometry and some Analysis. Also, Computational Complexity and some course about Computational Errors, for which you'll use a bit of Analysis.

And then discrete stuff, like Data Structures, Compilers, Linkers, etc. .

Most of the theory is essential not because you'll need it day to day, but because struggling and pushing through it, you get a better understanding and the ability to pick up new concepts, that you *will* need.

Imagine being a footballer and training for endurance through just running. Obviously running is not the same as playing football, but the endurance acquired will make you a better footballer than before.


Above all, don't worry about it. You'll be fine. You'll be taught everything from scratch, it's not hard, everything flows easily from one lecture to the next. *Always* attend lectures, attempt the exercises (even if you do them wring or are unable to come to a solution at all), and ask your professors questions in private. And team up with classmates to do homework together, especially if it counts towards your grade.

You can do it user.

>Avoid this path unless you want to compete for a data science job against hard science grads with advanced degrees who couldn't get a job in academia.
This one is somewhat true. If you're going for a data science job, you'll be mostly competing with people that did a Masters in Data Science + Bachelors in CS or Maths.
Usually, in their masters, the CS grads fill in on their maths skills, and the Maths grads fill in on their CS/programming skills.

That said, it's completely possible to pursue a data science job just with a CS bachelors, as long as you take the appropriate courses and do your dissertation on it, even better if you get a summer internship or research project (this one more doable) on something data science related.

Also, data science is not the only field you could go in obviously.

I barely use more than basic algebra, not even that most of the time.

>Learn databases or webdev.
There's a whole slew of things in between these that are enjoyable. "webdev" is not the only choice besides being a DBA.

Shit, as long as you're capable of learning a new language over the course of a few months and know one or two languages fairly well you'll be able to jump to another job without trouble if you have prior experience almost anywhere.

Got a thicc book for PHP and MySQL. Is this a good start?

this

I did CS, it was an applied maths course with a bit of actual programming, mostly in elective courses.

>A well paying job will require a BSc in Computer Scienc
>tfw not in CS I'm a junior and my school doesn't offer CS. RIP

You're doing it wrong. First you find the job you want, then you learn what they do.
Find something here stackoverflow.com/jobs

i really fail to see how math isn't useful in programming. knowing how to manipulate numbers in general will allow you to come up with more efficient and creative ways to solve problems.

>dissertation
Ain't nobody writing a dissertation for a bachelors. Go larp somewhere else fag

>highschool-tier math
>hurr gaymdev hard
Even the undergrad-tier meme learning requires more math than gaymdev. Now try formal verification.

why are so many people who are interested in being in this field also scared of math and asking how to evade it? I though math and computer science go hand in hand and everyone should love both

>Ain't nobody writing a dissertation for a bachelors.
??
Are you memeing or what?

Some Bachelors don't require it and some do. But in 90% of the time, you can write one in place of a module or 2. Unis have different names for it, some call it "research project", or "advanced topic" etc. .

I sure as hell wrote one (it wasn't mandatory, and
counted just for 1 module) and so has a sizeable chunk of my class.