I'm guessing there's a few old dogs on this forum. I'm going to guess a few of you probably work with embedded electronics in your day jobs. I'm interested in switching from Web Development to programming embedded systems.
Any advice you'd have for job searching and hints would be appreciated. I'm familiar with C/C++ and have burned eeproms in the past and know my way around a soldering iron.
J.
Embedded Electronics
Re: Embedded Electronics
I'm not sure what the salary of a embedded developer is these days but I did it in the past and I make a lot more money now as a Web developer now.
The big demand today seems to be Web Mobile IOS and Android apps.
The big demand today seems to be Web Mobile IOS and Android apps.
Re: Embedded Electronics
I think embedded is in quite big demand, although those who seek for such developers also have steep requirements, and you won't necessarily find the job too much amusing. These micros are used all around in more or less safety critical equipment, household stuff such as simple washing machines, then a huge chunk in process automation, factories, data acquisition, car electronics, then going steeper, rail, subways, medical stuff, and even higher, aviation and space. These all require micros since "modern" stuff can not be verified to work "proper", not even the hardware, much less the software, even including Linux if it comes to something with such steep safety demands like a car's steering or anything responsible for signalling or controlling the motion of railway equipment.
Be prepared to battle with lots of red tape in there. Documentation, formal requirements management, specification, testing and verification, all the likes. It can sap you dry even if your colleagues know well what they are doing, and a lot more if they don't. But these make our economy tick, which might be a motivation to still do these. And there are very few who can do these proper, so those are pretty much looked after.
Of course the entry to this world is nothing simple, but if you have the affinity for it, it is possible to find the way in step by step, likely entering through some company which does diverse things, along which there are project with modest safety requirements (such as household equipment). If you really know how to handle stuff in C, (and / or C++), and maybe even get to know some standards like MISRA including the proper use of said standard, then you may well find yourself eventually in the "deep water".
Be prepared to battle with lots of red tape in there. Documentation, formal requirements management, specification, testing and verification, all the likes. It can sap you dry even if your colleagues know well what they are doing, and a lot more if they don't. But these make our economy tick, which might be a motivation to still do these. And there are very few who can do these proper, so those are pretty much looked after.
Of course the entry to this world is nothing simple, but if you have the affinity for it, it is possible to find the way in step by step, likely entering through some company which does diverse things, along which there are project with modest safety requirements (such as household equipment). If you really know how to handle stuff in C, (and / or C++), and maybe even get to know some standards like MISRA including the proper use of said standard, then you may well find yourself eventually in the "deep water".
Re: Embedded Electronics
I good friend of mine worked for 15+ years in embed before moving to mobile dev on IOS and Android. I recall him saying he was sometime afraid to become unemployable because of the relatively low number of jobs available. And he was tired of working for small shops and being not so well paid. Mobile stuff is much more in demand and he seems pretty happy so far.
Re: Embedded Electronics
Huh... Here whenever I decide to look around to check who is hiring, I almost always see open positions at just about every larger firm dealing with anything safety critical, and these things won't just vanish in a puff of smoke. Maybe especially because mobile is the trend today, people are not that interested in learning working with the bare metal, and universities apparently neither have too many courses on the programming deal. I am not an electrical engineer, but an embedded programmer, a rare thing, something not even really covered by education despite that this niche, at least those who know the mechanics of programming in-depth, are in demand at such firms.
What I saw is that most electrical engineers assigned with programming duties simply can't really program, as in building larger applications in a traceable, maintainable manner. They are of course demanded to design and build the hardware, I really don't want to deny their skill by saying these, but they turn useless when it comes to building software adhering stringent standards. The assortment of commercial compilers for certain micros (such as the PIC) even show this: they frequently carry a few loathsome non-standard kludges to "fix" the language for people who don't really know the language from the start. The C language has a standard, but few know about it, and few are aware of the pitfalls of undefined or implementation defined behaviors for example. Still. Even after many decades of study on the matter. This then breaks software, introduces subtle bugs, unacceptable in anything safety critical. And this is only one facet of the complete story.
Of course I see that everywhere else with the ever decreasing prices making even things like this possible the use of small micros decrease everywhere where they can be exchanged to something of similar price capable to run a "normal" OS and software, but I think most of this change already happened. Where they still exist, they can't be exchanged due to the strict requirements of safety, so they will be around for many long decades still.
If you really want to see something on why and how this niche comes to be fundamental, you can read about the case of the Toyota Camry. Here is an easier to digest EETimes article, then, for those with dedication, the entire related court transcript can be found on the net. It is a dense but worthy read for anyone really interested in safety critical, how things can go utterly wrong in some decades of not really traceable coding once a piece of software becomes too complex to oversee in a few afternoons. Then there is for example the case of the 1996 Ariane 5 launch, more widely known since the software bug causing it turned out to be an easy to explain and understand overflow.
Of course this is definitely not for everyone as I mentioned above, but if you are in, you will find it is something quite demanded. Our economy just wouldn't work without these little buggers everywhere, you don't see them, yet they are there, doing their job in the dark confines of some important machinery, and you will only notice their presence when one makes it on the headline. And that's not nice.
What I saw is that most electrical engineers assigned with programming duties simply can't really program, as in building larger applications in a traceable, maintainable manner. They are of course demanded to design and build the hardware, I really don't want to deny their skill by saying these, but they turn useless when it comes to building software adhering stringent standards. The assortment of commercial compilers for certain micros (such as the PIC) even show this: they frequently carry a few loathsome non-standard kludges to "fix" the language for people who don't really know the language from the start. The C language has a standard, but few know about it, and few are aware of the pitfalls of undefined or implementation defined behaviors for example. Still. Even after many decades of study on the matter. This then breaks software, introduces subtle bugs, unacceptable in anything safety critical. And this is only one facet of the complete story.
Of course I see that everywhere else with the ever decreasing prices making even things like this possible the use of small micros decrease everywhere where they can be exchanged to something of similar price capable to run a "normal" OS and software, but I think most of this change already happened. Where they still exist, they can't be exchanged due to the strict requirements of safety, so they will be around for many long decades still.
If you really want to see something on why and how this niche comes to be fundamental, you can read about the case of the Toyota Camry. Here is an easier to digest EETimes article, then, for those with dedication, the entire related court transcript can be found on the net. It is a dense but worthy read for anyone really interested in safety critical, how things can go utterly wrong in some decades of not really traceable coding once a piece of software becomes too complex to oversee in a few afternoons. Then there is for example the case of the 1996 Ariane 5 launch, more widely known since the software bug causing it turned out to be an easy to explain and understand overflow.
Of course this is definitely not for everyone as I mentioned above, but if you are in, you will find it is something quite demanded. Our economy just wouldn't work without these little buggers everywhere, you don't see them, yet they are there, doing their job in the dark confines of some important machinery, and you will only notice their presence when one makes it on the headline. And that's not nice.