I own the outcome, from the first question to working software, directing AI across the build.
I've spent over a decade writing software. That experience guides how I use AI today. The projects below show what that looks like.
At a glance
- years building software
- 16
- personal projects built on my own
- 11
- teams led, from 2 to 7 people
- 5
Three projects, three ways of building

Pandacorp
AI-orchestratedA software factory I built and run with AI: 3 projects through it, and this site built and tested in 5 days.

Job Research
AI in the productAI job search for frontline workers. I led the app and built its pipeline, importing +1000 listings a day.

Alohub
Hand-builtOne editing engine for every section type. The framework I built reduced development time by 75%.
From the problem to working software
These are the decisions I pay most attention to.
Start with what matters
I focus the scope on the work that makes the biggest difference to the product.
Understand before designing
I understand the business and the problem before choosing how to structure the software.
Code the team can read
Clear names, small functions with one job each, and reuse of what is already there.
Ownership through to production
I take responsibility through to production. When a bug appears, I investigate the cause and fix it.
How I really use AI
I direct AI across design, code and review. I set the approach and decide what stays. Nothing ships until the tests, type checks and browser checks pass. If something fails, I investigate the cause before moving on.
See a full product I built this way: PandacorpAI does the work. I make the decisions.
The tools I build with
Got a problem worth solving?
That's the part I like best. Tell me what you're building.
Let's work together