I've been in IT for over 30 years. Networks, servers, helpdesks, MSP work, the whole spectrum. I know how technology works. I've just never been the person who built it from scratch.
That changed about a year ago.
The honest origin story
It started the way a lot of things start these days, I was messing around with AI tools and someone suggested I could use them to write code. I was skeptical. I'd tried to learn to code the traditional way a few times over the years. It never stuck. The gap between "I have an idea" and "I have a working thing" was just too wide.
AI closed that gap.
Not because it writes perfect code, it doesn't. Not because you can just describe a thing and it magically appears, you can't. But because it made it possible for someone with deep technical knowledge and zero formal development background to actually get things over the finish line.
I knew what I wanted to build. I understood the systems. I just needed a way to translate that into working software. Turns out that's exactly the gap AI is good at bridging.
What 30 years of IT actually gives you
Here's something nobody talks about: domain knowledge matters enormously when you're building with AI.
The reason I could go from zero repos to 70+ isn't that the AI did everything. It's that I already understood what I was building and why. I could spot when the output was wrong. I could describe problems precisely. I could make architecture decisions because I've been living inside technology for three decades.
A lot of people treat AI coding tools like a shortcut around understanding the thing. In my experience that's where it falls apart. The AI is a collaborator, not a replacement for knowing what you're doing. The more you bring to the table, the more it gives back.
What I've actually shipped
Apps for iOS. Desktop apps for Windows. Browser-based tools. Automation pipelines. A browser RPG. A memory system for AI assistants. Tools I built because I needed them, tools I built because someone I know needed them, and a few I built just to see if I could.
None of it would exist if I'd waited until I felt "ready" in the traditional sense. I'd still be waiting.
The thing I'd tell anyone on the fence
If you're sitting on an idea and telling yourself you'll build it when you learn to code properly, stop waiting.
The tools exist right now to take someone who understands a problem domain and turn them into someone who can ship a solution. It's not magic and it's not effortless. You'll get things wrong, you'll hit walls, you'll spend an afternoon debugging something embarrassing. That's just building things.
But the barrier between having an idea and having a working app has never been lower. If you've got the domain knowledge and the drive to figure things out, that's genuinely enough to get started.
I'm proof of that. 70 repos and counting.
Everything I build lives under the Axly's Customs banner, local-first, no telemetry, no subscriptions. More about what I'm working on at axly.com, and more of this kind of writing at axly.com/blog.