Kumar describes heart rate variability (HRV) as variation in the time between heartbeats and presents it as a signal of stress and recovery. His app compares recent HRV with a longer-term baseline to help users decide whether to take on work, keep to plan, or rest.
How to Optimize Your Health with React Native
Speaker
Tejas Kumar, an AI engineer at IBM and longtime software engineer, explains how he built a React Native health app with Expo to help users interpret heart rate variability and regulate their workload; the talk was hosted by Callstack, and the venue is not stated.
Kumar chose an iOS app because he needed access to HealthKit data, which a web app could not provide for this use case. He uses React Native to work with familiar JavaScript and TypeScript while accessing native features, and says it also makes sense when a team wants a mobile app without a separate web app.
The app presents four readiness states, from complete rest to going hard, using a traffic-light style interface. It compares a 60-day HRV baseline with a recent seven-day trend and also considers the coefficient of variation. Kumar demonstrated simulator data and added low HRV samples until the app recommended taking it easy.
Kumar chose Expo to manage React Native builds, upgrades, and deployment through Expo Application Services and TestFlight. He selected a HealthKit library by checking its recent commits, issue response times, and npm download trends. Because the app processes health data on the device, it works offline and does not need a separate database.
Takeaways
- Choose React Native when the web cannot access a capability your product needs, such as HealthKit.
- Evaluate libraries by maintenance activity, issue response, and download trends.
- Use mock data to exercise app states that depend on long histories, such as 28 days of HRV.
- Present complex calculations as simple guidance users can act on.
- Keeping sensitive calculations on device can reduce infrastructure and data storage complexity.
Mentioned in this talk
Search inside this talk
What was claimed 4
- role atTejas Kumar is AI engineer at
IBM“I work as an AI engineer at IBM.”
0:23inThe Critical Advantage With AIYouTube Verified3 sources· Talk
- previously atTejas Kumar previously at
Spotify
“I was previously at Spotify.”
YouTube Unverified1 source· Talk
- feature
Watson: Runs locally and is open-weight
“Our own Watson is a wonderful model that you can run locally. It's open source... It's open weights.”
YouTube Unverified1 source· Talk
- metricHRV: Time variation between successive heartbeats
“It's how much the time between each beat of your heart varies.”
YouTube Unverified1 source· Talk







