28 February 2026
What I Learned Building Software as an Enterprise Architect Using AI
Everything that made AI-assisted development work well turned out to be just good architecture practice. The skills I'd spent years building weren't a disadvantage. They were a significant advantage.
At ailíniú, we’ve always believed that great architecture is about alignment - connecting what a business wants to achieve with how it gets there. I’ve spent years helping organisations do exactly that. What I didn’t know, until recently, was whether that same thinking could help me actually build something - on my own, as a solo developer.
Late last year something shifted in the developer community. A bubbling up of excitement around AI-assisted development that felt different from what came before. I wanted to understand it - not just as an observer, but because my clients would eventually be asking me about it. So I did the only thing that made sense: I tried to build something myself.
The first attempt
It worked. That surprised me. Then I started poking at the edges and things broke quickly. The problem wasn’t the AI - it was me. I hadn’t provided enough structure or clarity about what I actually wanted. I worked through it incrementally and got to something functional, but it took far longer than it needed to.
The moment it clicked
Exploring how others were approaching AI-assisted development, something became clear. Everything that made it work well - clear scope, modular design, defined boundaries, structured thinking - was just good architecture practice. The skills I’d spent years building weren’t a disadvantage. They were a significant advantage. Developers were learning to think like architects. I already knew how.
What actually works
Break your project into small, clearly defined modules. Define your domains up front. Think about boundaries before features. And keep it concise - a small number of clear principles will serve you far better than a long verbose specification. The AI needs to understand what you want, not wade through a document that buries the intent in detail. This is, of course, exactly what good architecture has always been about.
What comes next
That experience gave me the confidence to go further. I’m now building three products - one addressing a gap I’ve observed repeatedly through my consulting work, another solving a problem in a sport I’m deeply involved in, and a third that’s entirely personal - something I built for my son. All more demanding, all a genuine test of everything I’ve learned. That’s a story for another post.
If you’re an architect or senior IT consultant who’s always felt the barrier to building was too high - I think that’s changed. The question is: what idea have you had sitting in the back of your mind that always felt just out of reach?
It might be closer than you think.