LAB · EXPERIMENTS · PROTOTYPES

I build to
understand.

Some ideas become products. Others exist simply to answer a question, test a technology or learn something new.

Build. Test. Learn. Repeat.
The lab is where ideas are allowed to still be unfinished.
WHY A LAB?

Not every idea needs to begin as a business.

Sometimes I build because I want to understand how something works. Those experiments often become the knowledge that makes later projects possible.

ON THE BENCH

Things I’m exploring.

01 IN DEVELOPMENT
Underwater robotics

DOE

Deep Ocean Explorer

A tethered underwater ROV built around cameras, sensors, propulsion and a modular architecture.

What does it actually take to build a useful underwater vehicle from accessible components?
02 ACTIVE PROTOTYPE
Game · Simulation

Sentinel

Systems through simulation

A space-game prototype used to explore systems, progression, resources, crews and decision-making.

How can simple systems produce interesting decisions?
03 EXPERIMENTING
AI · Architecture

AI Orchestration

Less waste. Better routing.

Experiments in model routing, orchestration, delegation and more efficient token usage.

Why use the most expensive model for every step if a smarter architecture can distribute the work?
04 BUILT
AI · Interaction

Prompt Builder

Structure before prompting

A tool designed to help turn an unclear intention into a structured instruction for AI.

Can we help someone communicate better with AI without teaching them all of prompt theory?
05 BUILT
Learning · AI

Words

Explain AI Words

A simple interface for explaining AI terminology without unnecessary jargon.

How do you make a jargon-heavy field understandable in seconds?
06 BUILT
Decision · AI

Capabilities

What can AI actually do?

A tool for thinking about whether a problem is suitable for AI and how much human review it requires.

Does the problem actually need AI?
THE PROCESS
01 QUESTION
→
02 PROTOTYPE
→
03 BREAK IT
→
04 LEARN
→
05 REBUILD
THE POINT

The prototype is not always the product.

Sometimes the most valuable outcome of an experiment is simply knowing what to build next — or what not to build at all.