Laughed off the deck — and why it was the best thing that happened

Before the demo, two small hardware changes made a big difference. I moved the sensor from a vertical orientation to a horizontal one — it's far more stable and much easier to design around. And I finally added a string guide, after watching the tether jump off the sensor wheel over and over. I genuinely can't believe it took me that long to think of it.

Then I called Sean from ASP, and he agreed to host a demo in four days. I scrambled to optimize everything I could and built a backup unit, because I had a feeling I'd need it.

Building software from a kitchen table

Part of that scramble happened away from my equipment. With no hardware to tinker with, I built out the software side instead: a React Native app for recording, Supabase for storing data, and a Railway-hosted API for processing. Building for iOS without a Mac is its own kind of pain — I couldn't simulate locally, so every build went to the cloud at roughly two dollars a pop. By the time things worked I was on my twenty-fifth build. None of it is wildly expensive, but none of it is free either.

The demo

Then I demoed at ASP, met some genuinely great people, and learned an enormous amount — mostly because so much went wrong. The sensor wheel had too much friction. A misprinted box started tearing apart. The line was too short and snapped once. Diving-block compatibility turned out to be a real question I hadn't thought through. And the part that still stings: I pulled out a fishing reel to retract the tether, got a good laugh from the room, and then the reel flew out of my hands and landed in a way that made the line saw straight through the plastic guide instead of riding gently along it.

The big takeaway was unambiguous: a proper retraction mechanism wasn't a nice-to-have anymore. It was a requirement. You don't forget a lesson the room laughs you through.