Coding as a career?

Hi

So I have tried Scratch, abit, on my raspberry pi and liked it.  Took a lot of concentration but was wondering with the rise of AI, if it was worth pursuing as a career.

I tried a legal career and couldn’t even get my foot in the door.  So taking later risk of upgrading my computer to do this is, worrying. 

Anyone got any advice?

Parents
  • If you do then learn something niche, like assembly.  Everyone and their mother can code in python, java, c++, but not many can code in assembly and even less can do it well.  It was one of the few things i enjoyed on a degree I I did in Wales some years back.  I think it was last year NASA was looking for someone that could write assembly for one of the probes they sent out long ago.  they wanted someone who was able to write code for a chipset that had quite limited ram, i think only a few kilobytes, so you had to be a bit of a guru, but they were offering good money for the job.

    If you learn python, java or php, well I would say dont bother.  The market is already saturated with programmers.  Also to get to the level at which a company would hire you is hard and takes a while.

  • Assembly code was part of my Master's in Electrical and Electronic Engineering and I found it very difficult, a lot harder than the high level languages. But because the degree with so wide with programming, maths, physics, business etc I didnt really have the time to properly have a go and focus on it. That is interesting to hear that NASA has an application for it

  • The old probes sent out long ago all ran on chipsets that used it, mainly due to the lack of ram and storage.  I think the one last year was a probe that was leaving our solar system and needed some final updates, so it was time sensitive on top, plus the chipset used was very specialised.  Apparently the people who built and configured had all since passed, so no one was left who could do it.  They published the requirements and software online in the hope someone would look at it, maybe learn it enough to get them out of the problem.  no idea if anyone did, but they were offering a decent sum for about 6 months work.

    As for difficulty, asm is difficult.  Not for the faint hearted.  Its not something I do anymore.  i never was any good at the higher level languages, but managed to pass that module and vb.  I can still debug code in python and php, but I dont write it.

  • I always strived to learn how they made chiptune scores for crack software.  They were often completely written in asm, but had full tunes that would run for minutes long, all in an executable that was often only 20-30k.  That was how i first got interested in it.  Well before University.  i did manage to make it play something after much faffing, but I was in awe at people who could write a chiptune symphony completely in asm.

  • I still do a bit of ASM coding now and then, though only for small time-critical routines within a wider framework using higher-level languages. There certainly are niches for it where microcontrollers with limited resources are used, or in the design of compilers/interpreters for higher-level languages. In my teens, I was much better at it, as it was the only efficient alternative to very slow BASIC interpreters on the old 8-bit machines that I started out with.

    The biggest problem is the lack of portability - moving from one chip architecture to another can be really hard as you often have to learn a whole new instruction set. I think that getting a little taste of it can be useful even if you don't use it in earnest, just to get a picture of what really goes on within the hardware and gain some understanding of why some algorithms might work more efficiently than others; though certainly not essential.

Reply
  • I still do a bit of ASM coding now and then, though only for small time-critical routines within a wider framework using higher-level languages. There certainly are niches for it where microcontrollers with limited resources are used, or in the design of compilers/interpreters for higher-level languages. In my teens, I was much better at it, as it was the only efficient alternative to very slow BASIC interpreters on the old 8-bit machines that I started out with.

    The biggest problem is the lack of portability - moving from one chip architecture to another can be really hard as you often have to learn a whole new instruction set. I think that getting a little taste of it can be useful even if you don't use it in earnest, just to get a picture of what really goes on within the hardware and gain some understanding of why some algorithms might work more efficiently than others; though certainly not essential.

Children
  • I always strived to learn how they made chiptune scores for crack software.  They were often completely written in asm, but had full tunes that would run for minutes long, all in an executable that was often only 20-30k.  That was how i first got interested in it.  Well before University.  i did manage to make it play something after much faffing, but I was in awe at people who could write a chiptune symphony completely in asm.