Challenge
SleepFit needed to be more than a sleep timer. It had to collect real monitoring data, structure it into meaningful sessions, sync it with a backend, explain it through AI, and motivate users to return.
Back to Projects
SleepFit
AI SLEEP MONITORING
An AI-powered sleep monitoring app that records the night, visualizes sleep patterns, and turns raw sensor data into personalized sleep insights.
Sleep monitoringSensor dataAI insightRewardsMobile UX
Solution
I built a full mobile and backend pipeline: sleep session recording, epoch-level sound and movement data, event detection, report visualization, Claude-based sleep comments, reward evaluation, and admin management.
Impact
SleepFit shows how mobile sensing, AI analysis, habit design, and retention mechanics can come together as a complete health app experience.
Sleep Product
Record the night. Explain the morning.
SleepFit is built around a simple promise: capture the night as data, then return it as something a person can understand and act on.
72sleep quality
LIVE SESSION
Noise + movement + events
stored as sleep epochs
Data Pipeline
A sleep app is only useful when the data becomes a story.
The important work happens between the start button and the morning report: sampling, classifying, syncing, explaining, and rewarding repeat behavior.
Monitor
Bedtime, wake-up target, permissions, and local session start.
Sample
Noise and movement are grouped into epoch-level sleep records.
Detect
Snoring, noise, waking, and turning events are stored with probability values.
Sync
Session summaries, epochs, and events are sent to Node.js and MSSQL.
Explain
Claude turns the recorded night into a personal sleep comment.
Reward
Goals, streaks, AI comments, and monitoring events become points.
Morning Review
Not just sleep time. Sleep context.
SleepFit treats the report as the product, not an afterthought. The user sees goals, patterns, events, and an AI comment that explains what the data suggests.
Real monitoring
The app records what happens through the night instead of asking users to remember it later.
AI explanation
Sleep quality becomes easier to understand when raw data is translated into plain-language feedback.
Habit loop
Rewards and point policies give users a reason to keep returning without making the app feel like a game first.
Retention Loop
The morning report has to bring the user back tonight.
SleepFit keeps the app story mobile-first: fast local session handling, synced analysis, and rewards that support repeated sleep tracking.
Local first
The session is saved on the device first, so the user can stop monitoring and see feedback quickly.
Synced later
Summaries, epochs, and events move to the backend without making the mobile flow feel heavy.
Rewarded habit
Points and goal events turn repeated sleep tracking into a habit loop, not a one-off report.
Mobile Surfaces
Designed for the quiet parts of the day.
The app has to feel calm at night, readable in the morning, and operationally complete behind the scenes.
Start
Begin a night session
Insights
Review recent records
Report
Open session detail
Rewards
Check earned points