Ask HN: Where do you see the future of computer science jobs? | Hacker News Reader

Ask HN: Where do you see the future of computer science jobs?

29 pointscardanome14 comments

As a classical Backend Web Developer I am a little bit worried about my field. Sure things are looking fine right now and there will always be some jobs in maintaining legacy PHP systems. Still wont hurt to keep my eyes open, right?

The classical Web is slowly dying. Lot's of people managed to transition into native App development for Android and iOS but how long will it last? Sure all the marketing people think that everyone and their dogs need an app to be trendy but do they really? Are people not sick of having to install another app? Isn't Facebook and Instagram all they care about?

AI is currently at the top of another hype cycle. It might have lasting staying power and real world applications this time. Heck it might even be able to avoid another AI winter but how will the churn of thousands of boot camp students chasing the next big thing effect the job market in the long term?

Personally I am really interested in Data Science so would be interesting to hear more about people working in this field. I heard it is a lot about cleaning up data which is something I actually like to do. Lot's of people are trying to get into the field though.

So what do you guys think? What are the best skills to learn to be ready for the future?

A startup launched here a few days ago that does autocomplete for the terminal. Not even all low-hanging fruit has been picked, let alone software for managing intricate processes deep within the economy or software catering to most niche interests.

The global SaaS market is growing at something like 10-20% year-on-year. Developer salaries have been rising steeply pretty much across the board. When the exponential growth ends, the steady state will likely be quite comfortable as well. You can pick whichever area you have an interest in and be safe for the foreseeable future.

BTW forget PHP systems, all the web and mobile apps built on top of 100s random NPM packages will require a tremendous manpower to deal with for decades.

> A startup launched here a few days ago that does autocomplete for the terminal.

Huh?

WebAssmebly/WebGPU/WebXR, without a doubt. You’ll be able to compile virtually any program from any language to run in the browser, this will enable a new breed of high performance web applications that just work across all devices, without having to go through walled gardens.

I second that. personally I really got into XR. I wanted WebXR specifically but it just isn't there just yet. maybe next year! for now I'm with Unity.

but generally anything WebSomethingSomething is a good starting point!

you will always need security - so anything cyber will do.

AI as a term is heavily and inappropriately used, but there is no doubt this field is solid. it just happens that everyone try to leverage it as a buzzword.

anything scientific is pretty secure as well. data science, analyitics, statistics, etc. some remarkably interesting (most must be boring jobs imo although I like the field in general)

and of course there's automation - he who controls automation controls the world. or at least and usually highly capable to generate passive income.

then there are semi-cs jobs, the computer part without the science, such as art although many design, graphics, gaming, etc. roles are pretty technical. entertainment industry ain't going no where that's for sure.

The biggest problem with software employment is missing fundamentals, poor training, and poor leadership. When employers advertise a software job, for example, they focus most importantly on language and tools. The focus is not about delivering a solution, despite all intentions to the contrary, but how to do a job (literacy).

The result of that systematic failure is an artificial perception of talent shortage. When people are unwilling to offer a solution developers instead turn to tools to fill that gap as though they can outsource experience and training for a product (typically free permissively licensed products).

Many software developers, as my experiences on HN suggest, are horrified to even address this problem and proposals are demonized. Every other industry has solved this problem. I suspect that if the industry is unwilling to address this matter it will take some catastrophic event and government will solve it for the industry.

Learning to handle data is probably solid. Data size and complexity will only grow. The science is by definition at the unexplored boundary. Have a solid understanding of the known and explore the boundary. but you don’t need to stess so much. Maintaining legacy systems is a never ending career.

Backend will always be around. Business need their data stored somewhere, and business logic applied to it. We've moved to providing APIs rather than sever side rendering but thats about it.

Data scientists are good at analysing the data. But they're not experts in encoding business logic in code.

Why do you think it's disappearing?

Web development is far from dying. Web3.js is as brilliant as the rise of jquery & Web 2.0.

I see more integration of packaged solutions/SAAS and as I seem to becoming an architect rather than a developer, maybe that's not so bad. I guess this has been happening slowly since I started in the 90s so it's not a massive surprise. As long as you're flexible and not tightly coupled to anything I'm pretty sure some kind of tech job is going to be available for a few decades.

Talking and explaining and solving business problems isn't going out of fashion.

Tangent. Do you actually mean native Android?

I've interviewed a couple times for Android dev jobs. The managers always ask if I've written native apps. I tell them I have written some apps in Java for Android, but haven't written any native apps. At this point they usually look confused or ask a for more details. Then I explain that actual native Android is C.

> Do you actually mean native Android?

I get your point and I am sympathetic to it but most people would consider solutions in Java or Kotlin to be native. It is simply a different use of the word "native". Honestly these days even solutions like React Native(!) and Flutter are called native.

So I would say if a manager type is asking you and you are doing more than pure web stack PWAs it is fair game to call it native.

It makes sense to clear it up when going for the interview. I don't want to be interviewing for a C job thinking it's Java.

Your answer would confuse me and is obviously confusing your interviewers. Maybe try a different answer next time?

Reply on news.ycombinator.com