Sleep Tight

Sleep Tight uses Codex and GPT-5.6 to hyper-personalize your bedtime routine from phone and watch signals, turning daily vitals into a safer light-and-sound wind-down plan.

Comment

Sleep Tight

Research gives us average sleep advice. Sleep Tight asks what happens when that advice becomes hyper-personalized to you.

Sleep Tight is a bedtime personalization tool for people who want a calmer, more targeted wind-down routine that adapts to their lifestyle. Each night, an Android phone exports the last 24 hours of available Health Connect vitals, while a Galaxy Watch captures a bounded 10-minute pre-sleep heart-rate window. At 10:05 p.m., Codex Scheduled analyzes the latest data, checks sleep quality where records are available, and chooses a safe background-sound and lamp-fade plan.

The experience is deliberately bounded: the lamp dims warmer and lower, background sound fades out, and the system avoids reacting to uncertain consumer sleep-stage estimates during sleep. Across repeated nights, Sleep Tight generates monthly learning logs that summarize what worked and how the next routine should be personalized.

Codex was central to the build, helping create the phone app, watch app, receiver, dashboards, lamp simulation, sleep report, tests, README, slides, and GitHub Pages walkthrough. GPT-5.6 shaped the product direction by moving the idea away from unreliable real-time sleep-stage intervention and toward pre-sleep signals that are safer and more actionable.

Built With

Share this project:

Updates