Projects
I am a generalist by choice. These are builds where the technology was new to me, the timeline was short, and the job was to lead a team to something that actually worked. What carries between them is not a stack. It is how fast I can learn one.
-
READI: coordinating the first 72 hours after an earthquake
Four people, 48 hours, and one question nobody could answer: when disaster strikes, what is the go-to tool for communicating and coordinating?
- Working end to end demo shipped and presented inside the time limit
- Model and front end built in parallel by separate sub teams, integrated on the first try
- 313 mapped zones across a 3 phase earthquake scenario, live on stage
-
Capital Compass: showing small business owners what money actually costs
I earned buy-in on the vision, built the solution independently for the first time alongside seasoned engineers, and had my design selected for us to fine-tune together.
- Third place, with my design selected from competing independent concepts
- Secured team buy in on the vision, then presented it to the judges
- Directed a full working build end to end with Claude Code, my first time doing it
-
Convos: a real-world conversation tool and social network
Convos helps you prepare for the situations you keep finding yourself in, and remembers what works for you next time.
- Working MVP shipped as an installable app
- 27 shipped iterations, each one user-facing, with the intake cut from four steps to three after early testing showed the fourth going unanswered
- Architecture built so the next phase, real accounts, a real cohort feed, a language model, costs a new layer rather than a rewrite
The tools change. The way I lead doesn't.
Scope calls made under pressure, structures designed so people can move in parallel, and unfamiliar tools learned fast enough to be useful by the deadline. The subject changes. The way I get to a result does not.