Psst... Lumi the AI assistant says hi! You're awesome for using accessibility tools.Skip to content
luinbytes.dev
<- Back to projects
Android AppLocal-first

Sleepr.

A quiet sleep companion for cycle-aware wake windows, optional live notifications, and private rhythm learning that stays on your phone.

PlatformAndroid
Cycle base90m
Cloud need0
ModeQuiet
01 / Wake Windows

Better mornings start the night before.

Sleepr focuses on the handful of bedtime decisions where sleep-cycle guidance is actually useful.

Plan from bedtime

Open Sleepr when you are about to sleep and get wake windows that land at cleaner cycle boundaries.

  • Wake suggestions based on sleep-cycle timing
  • Designed for quick decisions at night
  • Keeps the app focused on the next useful choice

Work around alarms

Use the next alarm as a real constraint instead of treating every night like a blank spreadsheet.

  • Surfaces guidance around the wake time you already picked
  • Helps explain when a morning may feel rough
  • Keeps the calculation understandable, not mystical

Adapt to real mornings

Rate mornings as rested or groggy so Sleepr can tune its cycle estimate to your lived pattern.

  • Morning feedback feeds personal cycle tuning
  • Gentle rating actions from notifications
  • Built for gradual learning instead of noisy daily micromanagement
02 / Ticker

A live hint when you want it.

Opt-in

The ongoing notification only appears when you enable it.

Minute-aware

Wake hints update with the clock instead of going stale in the shade.

Reboot-aware

If you opted in, Sleepr can restore the ticker after the phone restarts.

03 / Learning

Learns the quiet way.

Sleepr can learn from rough sleep and wake signals without turning your night into a logging chore. Screen-off, screen-on, and morning ratings become gentle evidence for better guidance.

Personal cycle estimate

The default 90-minute cycle is a starting point. Sleepr is built to move toward your actual mornings as it gathers enough local evidence.

Screen usage learning is opt-in through Android settings
Morning ratings keep the feedback loop simple
Cycle tuning improves slowly instead of overfitting one bad night
04 / Privacy

Sleep data should not become another account.

Sleepr is shaped around local guidance. The parts that know your schedule live on the phone.

Local-first by default

Sleepr's sleep profile, ratings, and learning signals are stored on your device. The core sleep model does not need an account or server.

Usage access is optional

Schedule learning can use Android's special usage-access permission to infer likely sleep and wake windows from screen activity. It is opt-in.

Notifications stay under your control

The live ticker is an explicit setting. If enabled, Sleepr can restore it after reboot; if disabled, it stays out of the way.

Built for quiet guidance

Sleepr avoids social feeds, leaderboards, cloud dashboards, and anxiety charts. The goal is a better next wake window.

05 / Build

Native Android, built for the phone.

Kotlin

Native Android application code

Jetpack Compose

Calm, responsive app UI

Room

Local database for sleep learning state

DataStore

Settings and onboarding preferences

WorkManager

Background scheduling and reminders

Hilt

Dependency injection across app services

06 / Android

Sleepr is coming to Android.

Built as a quiet local-first app for bedtime planning, wake-window guidance, and personal cycle learning.

Android appLaunch page in progress