Skip to content
BOBOXStudio

Android wellness app

Project Slowlight

A deliberately simple, offline visual breathing pacer built to make slow guided breathing easy to use at night without accounts, audio or distracting interface clutter.

Working prototype AndroidOfflineVisual pacingLow-distraction UX

The problem

The useful part of a breathing device should not require another piece of hardware.

The concept started from a simple observation: a gradually slowing visual rhythm can guide breathing before sleep, but the experience does not inherently need a dedicated physical device, an account, cloud services or a noisy feature set.

The insight

Strip the product back to the behaviour that matters.

The experience works best when it is visually calm, immediate and predictable. That meant removing analytics, ads, audio, network dependency and medical claims, then focusing on one job: show a clear breathing rhythm that progressively slows.

The build

Build a minimal offline Android experience around the breathing session.

The prototype uses a full-screen expanding and contracting visual pacer, a gradual ramp from roughly 11 to 6 breaths per minute, 8 and 20 minute session options, brightness control and basic pause, resume and stop behaviour.

Working on-device

The prototype was built, installed and used on a Samsung S24 Ultra rather than remaining a design concept.

Offline by design

No accounts, analytics, ads, cloud dependency or network access are needed for the core session.

Restrained interface

Flat surfaces, subtle borders and minimal controls keep the experience intentionally quiet rather than visually performative.

Clear product boundary

It is positioned as guided breathing for relaxation rather than making medical or insomnia-treatment claims.

Why it matters

Sometimes good product design is mostly deciding what the app does not need.

Every project is different, but the goal stays simple: understand what needs to work, design it clearly and ship something useful that can be maintained after launch.

Have a problem worth building around?

Start a project