Building fully-functional HOA management software with modern AI tools — for my own neighborhood.

My Role

Designer and builder. I used the latest AI tooling to design and ship the whole thing myself.

The Problem

My HOA board needed the right technology for the money it had. I started as an evaluator rather than a builder: I sat in on the board’s problems while vetting existing HOA-management vendors, and came away convinced none of them fit what this board actually needed. So we built our own.

Constraints

Real users — my own neighbours — with real requirements and no tolerance for a half-working tool. A solo build. And a deliberate second agenda: this was my exercise in genuinely mastering AI-assisted development rather than dabbling with it.

Process — buy first, then build

The vendor evaluation was the design research. Working through what commercial HOA software does and does not do, while listening to the board describe its month, told me more about the requirements than a greenfield discovery process would have.

From there I designed and built the full platform — dues, rules, vendors, messaging, voting, and more — using AI tools across the entire workflow, from first sketch to shipped feature.

Key Decisions

Designing to the board’s actual needs, not the category’s feature checklist. Every feature was grounded in something the board genuinely wanted, then designed for the most usable experience I could manage — rather than matching what vendors happened to list on a comparison page.

Outcome

A working platform in active use by the neighborhood.

The detail I am proudest of is an in-app bug and feature-request flow that routes straight into a development agent. It can find the bug, fix it, and automatically email that specific resident when the fix ships.

Reflection

A genuine chance to master the AI tooling I now use everywhere else.