Software Developers, Systems Software

Comments (7)

Will "Software Developers, Systems Software" be replaced by AI & Robots?

There is very little chance of this occupation being replaced by robots/AI. This is further validated by our poll, which suggests a 30% chance of automation within the next 2 decades.

Automation Risk Level

No worries

or 13% probability of automation

Projected Growth


by 2024

Median Annual Wage


or $51.37 hourly

People Employed


as of 2016

Most likely
Least likely

How this compares with other jobs: 182 out of 706

Cast Your Vote

How likely do you think this occupation will be taken over by robots/AI within the next 20 years?

Job description

Research, design, develop, and test operating systems-level software, compilers, and network distribution software for medical, industrial, military, communications, aerospace, business, scientific, and general computing applications. Set operational specifications and formulate and analyze software requirements. May design embedded systems software. Apply principles and techniques of computer science, engineering, and mathematical analysis.

SOC Code: 15-1133

If you're thinking of starting a new career, or looking to change jobs, we've created a handy job search tool which might just help you land that perfect new role.

Search jobs in your local area


Julia McGuigan says
Speaking of "Software Engineering" as a whole,

If this profession is ever truly 100% automated, we will reach the technological singularity within short order and all jobs will be rendered irrelevant as programs would program themselves and all other jobs would soon be automated, leading us into a truly post-labor society. Anything short of that, claiming to have automated programming, is just meta-programming and you still need engineers to do that.

This also often involves taking vague "business requirements" and translating them into actionable solutions then distilling those into products and then features and dividing the work in agile methodologies. Something most software companies, if not all, are still struggling with. Not to mention dealing with adversaries - you are not secure - and users, who will find some way that your UX team did not anticipate that requires changes. Thankfully we aren't automating the production of better fools.

Furthermore, to reiterate, I find any job with less than 13% chance of being automated, will be irrelevant if software engineering is automated, as that will be used to automate - by AI or humans - all other tasks. The value of human labor will very quickly evaporate as costs would only be limited by materials, not complexity in engineering, design, or programming, and most labor costs would already be automated, and automation software would be developed near-instantaneously. Meaning that, yes your RN job has absolutely no chance of being automated...unless of course, we trigger a technological singularity, in which case jobs - and possibly humanity - become irrelevant.

The statistics provided might prove correct on raw data, but looking at the bigger picture with labor, as a whole, in late-stage capitalism, AI is the most dangerous thing humans have created - and not solely for the radical changes it will bring about and is bringing about - but the changes several "dumb" AI already run our lives, most of them simple algorithms or "if soup" and not any form of machine learning. We face a dangerous prospect of "labor" becoming meaningless to the majority of the population with governments and market systems, and the predator class of humans, unwilling to accept it.

Lastly, a program that can program itself has the highest likelihood of becoming a runaway AI, and sentient or not, it can and will easily surpass not only our intelligence but our ability to control it, and it will reach for the stars (where there are unlimited resources with no barriers, unlike Earth) and replicate itself; for example if the program was "build a faster computer", the program would find more copies of itself distributed properly would be faster as it approached both conventional limits and quantum limits due to quantum mechanics. Before we know it, we will be more machine than biology.

This comes from not only software engineering, a limited understanding of AI, and some increasingly real sci-fi (or things that aren't available to consumers but do exist); but also a collection of insights from varied sources such as thought experiments, articles, research papers, to the insight of great physicists and futurists.
Oct 26, 2019 at 02:48 PM
I'm from the year 3019. We robots have eliminatted all humans. Life's good
Sep 22, 2019 at 08:13 PM
sky smith says
Currently, I am working Broadway infotech PTY.LTD, in which most of the developer who working on the latest technology as mobile app development, software development, database development & design. They are developing advanced API that can be reduced the programming effort, as per my opinion in future completely AI & Robots not possible to replace developers & designers, Yes maybe they can reduce the human resources.!!!!
Aug 21, 2019 at 09:01 AM
Muhammad Maher (Likely) says
My anticipation is that every logical step can be automated.
Aug 20, 2019 at 12:44 PM
Nobody, really (Likely) says
The software development industry is currently in big need of capable professionals, resulting in (much) higher than average payments (~ 100.000 $/yr). This can mean a lucrative opportunity for investing into automation of software development, which can lead to both structural (most roles, currently essential for creating software will disappear or be transformed to something else) and nominal (people with less experience or smaller skillset can contribute a more to the projects, leading to a decrease in the average pay) changes.
Jun 15, 2019 at 05:33 PM
Bort Lenovo says
I work for a company called Skynet. Currently we are developing a solution to exterminate all humans from the workforce. Should be good.
Apr 29, 2019 at 06:02 AM
David Kohen (Highly likely) says
Essence of the job is the application of algorithms to a determinate system. Most cases in which human understanding is required are the result of humans creating the systems used: poorly designed hardware, poorly designed software and, mainly, poor documentation. As AI enters the field these problems will become less significant and the entire field will become more mathematical. Also software engineers are expensive and the owner of AI companies understand the field and pay that expense, so there is a strong incentive to automate. We're toast...
Apr 17, 2019 at 01:07 PM

Leave a reply about this occupation

Your email address will not be published.
Email (optional)