FilmedSurf

App icon - 1024 x 1024, uncropped

Inspiration

I surf. And like almost every surfer I know, I had never once seen myself surf.

Surfing happens out in the water, on your own, which makes it one of the hardest sports to film. But from the line-up I kept noticing the same thing: there is always somebody on the beach. A partner waiting with a towel, a friend, someone walking a dog, another surfer sitting out the tide. Every one of them already has a camera in their pocket.

The footage does not exist because nobody has a reason to press record. Give the people who are standing there anyway an incentive, and the problem solves itself. So I built the thing that sits between the surfer in the water and the person on the sand — a marketplace where one films and the other buys. No special kit required: a phone is enough, and the app has its own camera built in.

What it does

FilmedSurf has two sides. Surfers want to find footage of themselves. Filmers want to sell what they shot. The app is the place those two finally meet.

Logging a session is two taps — paddling out, and coming in. Session, spot, conditions and time are written for you, and it works from an Apple Watch, so it survives wet hands and a wetsuit.

Then your wave finds you. Any clip whose shooting window overlaps your logged session at that break turns up under "might be you". The match is place x time — never faces. We run no face recognition, and the filmer never has to learn who you are. What turns up plays watermarked; buy it and the clean file lands in your camera roll.

One frame is worth keeping. Stop the clip on a single frame and you can save it for nothing, turn it into AI art, or take it through to our own store and have it made into something you keep — a poster, a canvas, a mug, a tote, a shirt.

Beach clean. Photograph the rubbish you carried off the beach. A vision model checks it against four conditions — near the sea, shot today, taken with the camera right here, and actually rubbish — and then the photo is thrown away. We keep the credit, not the picture. Every accepted clean-up buys you one more go at the AI art.

How we built it

Expo / React Native, a Hono + Postgres API on Fly.io, Stripe Connect for the marketplace, RevenueCat for the in-app purchase, and a Shopify storefront on a print-on-demand partner for the physical goods.

I did not write a single character of this code by hand. All of it was built with Claude Code. It is also my first mobile app. Idea, design, payments, App Store release — one person, no team, no prior iOS experience.

Challenges we ran into

Going international. Seven countries at launch, each with its own currency and its own payout rules. Our first beach-clean reward was a discount on your next clip — it broke down outside a single currency, and it only ever paid out to people who were already buying. We replaced it with free AI art tries, which work on your own footage in every country.

Fitting an in-app purchase into a marketplace. Apple's rules and Stripe Connect's rules point in opposite directions. Working out what belongs in an IAP (AI art made from your own frame) and what belongs on the filmer's own Stripe account (someone else's clip) took longer than building either one.

Face blur. Deciding who needs blurring, when a human has to look, and making the safety gates fail-safe instead of fail-open — so that turning blur on can never quietly let something through unreviewed.

Apple Watch. Session state and GPS on a watch, with the phone left in the car park, is a different engineering problem from the same feature on a phone.

Accomplishments that we're proud of

Shipping it. A first mobile app, taken alone from a thought in the water to a live App Store listing.

A marketplace that finds your wave without ever matching a face.

And a free half that is not a loss leader: the surf log is the only route by which footage can reach a person, so the free feature is precisely what makes the paid one possible. Monetization and a contribution back to the surf community — the beach clean — were designed in from the start, not bolted on at the end.

What we learned

The design decisions you make at the very beginning matter more than anything you do later. Every hard week this summer traced back to something settled in the first fortnight — and the choices that held up were the ones I could state in a single sentence.