Point your phone at the scene. Framester reads the frame in real time — subject, horizon, headroom — and tells you, in one sentence, how to fix the shot before you tap the shutter.
Tilt down a touch — barely any space above her head.
Modern iPhones ship a sensor and an image pipeline that will out-shoot most DSLRs from a decade ago. The hardware is not the bottleneck.
The bottleneck is the moment before the shutter — the two seconds where you point the phone, don't think about where the horizon is, don't notice the subject is dead-centered, and tap.
Framester is a teacher sitting over the viewfinder. It won't touch the ISP. It won't apply a filter. It just quietly tells you how to stand half a step to the left.
Vision reads the scene. A rule engine evaluates it. Apple Intelligence speaks the fix. Nothing leaves your phone.
Apple's on-device framework detects faces, body pose, horizon, and salient subjects on every analyzed frame. Runs on the Neural Engine. Thermally aware.
Pure Swift. Takes Vision output, scores the frame against three composition rules, decides whether the overlay nudges you or stays quiet.
Apple Intelligence's Foundation Models turn the composition state into one short imperative sentence. Rate-limited, hysteresis-gated, cancellable.
Tilt down a touch — barely any space above her head.
Framester runs on iPhone 15 Pro and later, on iOS 26 or newer. It needs the Neural Engine for Vision and the Foundation Models runtime for the coach. There is no fallback branch for older hardware.
This isn't a disclaimer. It's the whole point. The coach line is the product — and the coach line is Apple Intelligence running on-device, in real time, next to the viewfinder.
No newsletter. No drip. We send one message when Framester is live in the App Store, and then we're done. Promise.