Back to Work

Meter Burn

A self-coached training app for iPhone, iPad and Apple Watch — built, signed, reviewed and published to the App Store by the studio.

On the App StorePublicly listed and cleared through Apple's review — a distribution bar an unsigned download never has to meet.
3 devices, one buildiPhone, iPad and a standalone Apple Watch app that runs with no phone present, shipped as a single bundle.
Zero data collectedNo account, no analytics, no third-party SDKs. Training data stays in the athlete's own iCloud and Apple Health.

The problem it solves

Training apps tend to hand you a plan written for a stranger, or leave you doing arithmetic between sets. Meter Burn does neither: the athlete authors the programming, the app reads the log, and when an exercise has earned a step it proposes one — and then waits. Nothing changes unless the athlete agrees.

Why it is proof

It is a Swift application whose domain logic lives in a headless package tested without a simulator, behind a release pipeline that makes sixteen local assertions that the archive is genuinely distributable — and refuses to ship a data model that the production CloudKit schema has never heard of, the one failure no build-time check can see. Shipping to a reviewed store means the verification has to hold in public.

Handling health data without collecting it

The app reads heart rate, active energy and — only if the athlete switches it on — body composition from Apple Health, displays them, and stores none of it. Its App Privacy label reads “Data Not Collected”, which is a claim about architecture rather than policy: there is no server to send anything to.

The proof point

A regulated, reviewed distribution channel and health data handled correctly — in something the studio ships itself, where anyone can download it and check.