Skip to content
All work

Nailstry

Nailstry — app screenshots

Overview

Nailstry is an e-commerce app for finding and buying press-on nail designs. Its standout feature is AI nail measurement: instead of guessing your size from a chart, you measure your nails with the camera and order a set that actually fits.

My role

I owned the AI nail measurement feature, and improved the performance and reliability of the wider app along with the Apple Pay checkout.

Hardest part

A photo has no sense of scale. Two people can shoot the same nail and get wildly different pixel sizes depending on distance and angle, so a measurement in pixels is worthless on its own, and a wrong size means a return, which is the one thing an e-commerce app can’t afford.

I solved it by calibrating every measurement against a known reference in frame, so nail width could be converted into real millimetres rather than pixels, and by guiding users into a consistent camera position so shots stayed comparable.

Architecture & stack

Swift on iOS, with computer vision driving the measurement flow, FirebaseML for the ML side, and Apple Pay handling checkout.