Why I’m Still Learning to Code Even With AI

Source: Dev.to

Introduction

A while ago, I caught myself wondering:

“Am I too late?”

Not because I started coding late, but because every week there seems to be a new AI demo building apps, writing features, fixing bugs, and explaining code faster than I can even think through the problem. When you’re still learning, that can get in your head.

You start asking yourself:

If AI can already do this, what exactly am I trying to learn?

I’ve been there.

The Challenge

I’m still early in my journey. I’m building projects, learning Laravel, breaking things, fixing them, and slowly getting better at understanding how software actually comes together.

Lately I’ve been working on a team task management app. At first it felt straightforward:

• A team has members.
• Owner can invite other members.
• Tasks can be created and assigned.

Simple—until it isn’t.

While working on the invitation feature, I started moving logic out of controllers and into service classes, added form requests, then policies. Suddenly I realized something bigger than the feature itself.

The hard part wasn’t writing code.
The hard part was deciding what code should exist.

What AI Can and Can’t Do

AI can absolutely generate code, and sometimes it produces surprisingly solid snippets. However, while building that feature, AI wasn’t the one deciding things like:

• Should this logic live in the controller or a service?
• What happens if someone sends the same invitation twice?
• Who should be allowed to invite people?
• What happens when someone leaves a team?
• How do I structure this so future me won’t hate present me?

Those decisions were still mine, and that’s the part I actually want to learn.

Shifting Mindset

When I first started coding, I treated code like a puzzle—adding pieces until the feature worked, then avoiding it for fear of breaking something else. Lately, I’ve been moving away from that mindset, paying more attention to system design—not in the “big tech interview” way, but in a practical, intentional way.

I now ask myself not only:

“How do I make this work?”

but also:

“Why should it work this way?”

This shift has made coding feel different—and more interesting. The more I learn, the more I realize coding isn’t mainly about syntax; it’s about trade‑offs, structure, and deciding what belongs where. It’s about understanding the problem well enough to build something that won’t collapse the moment you add a new feature.

Learning Takeaways

AI doesn’t discourage me; it clarifies what actually matters.

• AI can help me write code faster.
• AI can explain concepts and compare approaches.
• AI can point out mistakes I didn’t notice.

But AI still doesn’t replace judgment. It doesn’t automatically know the trade‑offs inside my project, the shortcuts I took last week, or what future me will struggle to maintain. That part still comes from learning.

If you’re just starting—or thinking about starting—and AI makes you wonder if coding is still worth learning, here’s my honest take:

This might actually be one of the best times to start.

Now you can get unstuck faster, ask better questions, and explore more ideas in less time. The real learning still happens when you decide:

• what to build
• how to structure it
• which trade‑offs you accept
• what code stays
• what code needs to be deleted

That part still belongs to you.

Conclusion

I’m still learning too:

• Still writing code I later refactor.
• Still realizing how much I don’t know.
• Still trying to go from “it works” to “this makes sense.”

And right now, that feels like progress.

AI didn’t make me want to stop learning; it made me realize something more important: the value isn’t just in writing code, but in understanding why that code should exist at all.

Note: I use AI while learning, not to skip the work, but to understand faster, explore alternatives, and challenge my thinking.