There’s a particular feeling that comes with coding.
Not the sterile version people imagine — lines of syntax and error logs — but the quiet, almost creative rhythm that emerges when an idea turns into something that works.
I felt that again this weekend.
After switching to Mac, there was one small thing I couldn’t find: an app that let me browse picture folders for project files with the kind of file-explorer experience I actually wanted. Nothing complicated. Just something that felt right from a workflow perspective.
So instead of continuing to search for the perfect tool, I tried something else.
I opened Xcode.
Turned on Codex.
And started building it myself.
Within a couple of minutes, I had a working prototype.
Not a theoretical prototype. A working tool that did exactly what I needed.
Then the interesting part began.
Instead of long development cycles, I found myself experimenting in rapid bursts. Prompting behaviours. Tweaking keyboard navigation. Adjusting little pieces of user experience. Trying an idea, seeing it appear almost instantly, refining it again.
Prompt.
Test.
Refine.
Repeat.
Within an evening it started to feel like a miniature release cycle happening in real time.
And I have to say — I’m impressed.
Right now the app is simple. It serves a specific purpose for me. But with a bit more time actually learning the code behind it, doing some proper QA, and cleaning up the structure, it might even become something worth publishing just for the fun of it.
What struck me most, though, was the contrast with how things used to be.
Back in high school in the mid-2000s, building something like this was a completely different experience. Learning the language alone could take weeks. Writing an app during school holidays might turn into a month-long project once you factored in documentation, trial-and-error, and the slow grind of figuring out why something wasn’t compiling.
Progress was linear. Slow. Hard-won.
Now it feels different.
The barrier between idea and implementation has collapsed.
You still need to understand what you want to build — maybe more than ever — but the mechanics of building it have accelerated dramatically. It’s less about fighting syntax and more about shaping intent.
Which raises an interesting question about where this all goes.
For decades, software development rewarded people who could memorise systems and execute instructions precisely. The craft was partly creativity, but also deeply procedural.
Now the landscape is shifting.
If tools can increasingly handle the mechanical parts of coding, the value may move further upstream — toward imagination, problem framing, and design thinking. Toward people who can describe what should exist, not just implement it line by line.
In other words, toward creativity.
Maybe the future developer looks a little different from the one we grew up imagining. Less keyboard technician, more architect of ideas.
Or maybe it’s something in between.
For now, I’m just enjoying the vibe of it again — that feeling when you build something small, useful, and unexpectedly fast. The quiet satisfaction of watching an idea become real in a few prompts and a bit of curiosity.
And every now and then I wonder:
If this is what coding feels like today, what will it feel like in another ten years?