My tablet-focused fitness app is already 80 % built and communicating with a Bluetooth FTMS-enabled kayak machine, but the live metrics coming through are wrong. I need someone who truly understands the FTMS specification to dig into the data stream, identify why cadence, power, distance and other values are being mis-parsed, and deliver consistently accurate readings on both iOS and Android. Where we stand • Core Bluetooth connection logic is in place and stays connected, yet the payload is interpreted incorrectly. • UI screens and navigation are mostly there, though some components still use placeholder graphics and layout. • The backend (user accounts, workout history, cloud sync) is wired up but several endpoints are still mocked. What I need from you 1. Correct extraction and mapping of all FTMS characteristics for the kayak machine so numbers match the console in real time. 2. Polished UI/UX for the remaining screens: workout summary, settings, and the live dashboard once the data is fixed. 3. Final backend integration: swap the stubs for the live API calls, test upload/download of sessions, and harden error handling. 4. General bug fixes uncovered during the above work and a quick regression pass on both platforms. Acceptance criteria – Real-time metrics within ±1 % of the machine’s own display over a 30-minute test. – No crashes or memory leaks flagged by Xcode Instruments or Android Profiler. – Successful login, workout upload, and history retrieval for three consecutive sessions. – UI matches the provided Figma file across the target tablet resolutions. Tech stack in play: native Bluetooth LE / FTMS, Swift & Kotlin (a small shared C++ parser for FTMS frames), Firebase for auth and storage. If you have better tooling suggestions I’m open, but deep FTMS expertise is non-negotiable. Code is in a private Git repo; I’ll grant access once we agree on milestones. Looking forward to getting this across the finish line quickly and cleanly.