Full-stack 開發者轉型 DevOps：AI 是否改變了我對軟件開發的看法？

From Developer to DevOps: Did AI Change the Way I See Software Development?

Recently, I made a pretty significant change in my professional career.

I switched sides.

I went from building applications, designing solutions, and turning ideas into code to working closer to the infrastructure that keeps those applications running in the cloud.

I'm now working as a DevOps Engineer, and although I still have a developer's mindset, my day-to-day work looks quite different.

But what made me take this step?

When AI changes the way you work

There's no denying that AI has dramatically changed the IT industry. Tools that once seemed experimental are now part of our everyday workflows, helping us write code, debug issues, generate tests, and even design entire features.

I've been using AI tools in my development workflow for quite some time, and they've undoubtedly made me more productive in many situations.

However, at some point, I started feeling that something wasn't quite right.

Maybe it was the way I was working. Maybe it was how I had integrated AI into my daily workflow. Or perhaps it was simply the natural evolution of an industry that's changing faster than we're used to.

Whatever the reason, software development was starting to feel different from what I had originally fallen in love with.

Are we becoming reviewers instead of developers?

One of the biggest questions I've found myself asking is:

How long until my role becomes unnecessary?

It's a difficult question, and I don't think anyone has a definitive answer.

AI can already generate surprisingly good code, solve complex problems, and handle tasks that used to require hours of manual work. As these tools improve, it's natural to wonder what the future holds for software engineers.

Will software development disappear? I don't think so.

Software is much more than writing code. Understanding business requirements, making architectural decisions, ensuring security, maintaining systems, and taking responsibility for the final product are all essential parts of the job.

AI may change how we approach these responsibilities, but that doesn't necessarily mean they'll disappear.

Still, I think there's an important distinction between using AI to improve the way you work and feeling that you're gradually losing the part of your work that made you enjoy it in the first place.

For me, that distinction became increasingly important.

I started wondering whether I was still enjoying the process of building software or whether I was spending more time supervising the process of having software built for me.

And that question made me reconsider my career path.

A different side of IT

Rather than waiting to see how things would evolve, I decided to make a significant change.

I moved into DevOps.

Now, I'm working closer to cloud infrastructure, containers, Kubernetes, CI/CD pipelines, and the systems that allow applications to run reliably and at scale.

It's a different perspective on the same ecosystem.

Instead of focusing primarily on building the application itself, I'm learning more about how to deploy it, operate it, automate its lifecycle, and keep the underlying infrastructure reliable.

And honestly, that's exciting.

There's still plenty to learn, plenty to automate, and plenty of complex problems to solve. My development background hasn't become useless; if anything, it gives me a different perspective when approaching infrastructure and automation challenges.

After all, infrastructure is increasingly managed through code, pipelines, configuration, and automation. The boundaries between software development and operations aren't as clear-cut as they once were.

Perhaps I'm not leaving software development behind as much as I'm exploring another way to practise it.

A developer at heart

I don't regret my time as a developer. Quite the opposite.

Building applications taught me how to break down problems, design systems, understand requirements, and think about the people who ultimately use the software.

Those skills are still valuable, even if I'm applying them in a different context.

I also don't believe that moving into DevOps is a way to escape AI. AI is transforming infrastructure management, cloud operations, monitoring, and automation too.

Instead, this change is about finding a professional path that motivates me, challenges me, and gives me room to grow.

I'm still the same person who enjoys solving technical problems and building things. I've simply changed the layer of the stack I'm focusing on.

And who knows? Perhaps my development background will become one of my greatest strengths in this new role.

So, was it the right decision?

I don't have a definitive answer yet. I'm still at the beginning of this new chapter, and there's a lot to learn.

But I do know that sometimes, making a career change isn't about running away from something. It's about moving towards something that excites you again.

And that's what I'm trying to do.

Now I'd love to hear your thoughts:

Developers: Has AI changed the way you feel about your work?
DevOps and platform engineers: Do you think having a development background gives you an advantage?
Everyone in IT: If you were starting your career today, would you still choose software development?

I'm genuinely curious to hear different perspectives. Let's discuss it in the comments!