Infolitz

Problem

‍

High packet-loss in the companion app
The first-generation Flutter app was dropping BLE packets coming from the RF glucose sensor. Gaps in the glucose curve made it impossible for study monitors to validate accuracy against lab draws.

‍

Evolving payload schema, static parser
The wearable firmware team had added new calibration bytes and a version flag, but the mobile app still assumed the original byte frame.

‍

No field-ready diagnostics
When dropouts appeared at trial sites, support staff had no built-in log or trace to share—only screen recordings. Reproducing issues required shipping devices back to engineering, delaying root-cause analysis.

‍

Solution

‍

Stronger, more reliable data channel: Connection parameters were tuned and lightweight confirmation logic added so the app now detects missing packets and recovers them automatically

Comprehensive code & workflow audit: We began with a deep dive into the existing Flutter code-base, BLE settings, and firmware packet flow to pinpoint where losses and crashes originated.

On-device rotating log: Implemented a circular, encrypted log (10 MB) capturing BLE RSSI, sequence gaps, parsing errors and user actions.

‍