We Have Always Built on Layers. This Layer (AI) Is Just Smarter.

There’s a lot of concern right now about AI tools doing junior developers a disservice. That they won’t learn fundamentals. That they’ll become dependent. That we’re skipping something important.

But this fear isn’t new. It shows up every time a new abstraction layer is introduced.

We’ve always built on top of layers.

First it was punch cards and physical wiring. Then machine code and assembly. Then higher-level programming languages. Then operating systems that handled memory and processes. Then libraries and frameworks. Then managed runtimes. Then cloud infrastructure. Then platform ecosystems.

Each step removed a kind of manual work that once defined “real engineering.” And every time, people worried we were making developers weaker.

What actually happened was the opposite. We made them capable of solving bigger problems.

Progress happens when we automate yesterday’s complexity so we can focus on tomorrow’s problems.

That’s not a new idea. It’s the story of software itself.

Abstractions don’t erase understanding. They shift where understanding lives. Instead of spending all our energy on mechanics, we spend it on design, systems, tradeoffs, and impact. The work moves up the stack.

AI is just the next layer.

Yes, the people who are naturally curious, who want to understand what’s happening under the hood, who enjoy thinking in systems, will still go deeper. They always have. AI doesn’t remove that path. It actually gives them more surface area to explore.

But we shouldn’t confuse curiosity with suffering. Not everyone has to start at the bottom of the stack to be capable or valuable.

What matters is not whether someone wrote memory allocators by hand or built everything from scratch. What matters is whether they can:

• reason about systems
• make thoughtful tradeoffs
• understand user impact
• build responsibly
• and keep learning

Working yourself out of a job

There’s also an old idea in engineering that feels especially relevant now: the job of a programmer is to work themselves out of a job.

Not by becoming unnecessary, but by making systems simpler, safer, and more automated so people can move on to higher-order problems. That’s what progress has always looked like.

Compilers did that. Operating systems did that. Frameworks did that. Cloud did that.

AI is doing the same thing, just faster and in a way that touches thinking itself.

Being afraid of AI is a little like being afraid of compilers because assembly felt more “real.” Or being afraid of cloud because managing servers made you feel powerful. Each generation had its version of this anxiety.

SaaS didn’t disappear when platforms emerged. Strong platforms became more valuable. The same thing will happen here. Strong engineers won’t disappear. Their work will shift.

So no, AI isn’t doing juniors a disservice. It’s giving them a different starting line.

They won’t be weaker. They’ll just be standing on a higher layer of the stack.