Companion App Modernization & Firmware Optimization for A U.S. HealthTech Wearables Startup
Our Clients
Engagement Overview
A San Francisco–based healthtech startup built clinically validated wearable patches that tracked heart rate variability (HRV) and delivered real-time breathing exercises to reduce stress and anxiety. The product had strong clinical backing and partnerships with UCSF and Stanford, but reliability issues in the companion app, BLE connectivity, and firmware pipeline made it hard to scale confidently.
TechBlocks came in as the end-to-end engineering lead. We modernized the companion app, stabilized BLE and OTA firmware workflows, and rebuilt the QA pipeline so the client could support over 10,000 daily users with a reliable experience, faster releases, and production-grade observability.
Key Challenges
The platform delivered real clinical value but struggled to behave like a stable, production-ready system.
The companion app crashed frequently, BLE connectivity was unpredictable, OTA firmware updates were risky, and sensor data was difficult to interpret. These issues drove up support costs, slowed releases, and threatened user trust at scale.
Unreliable Mobile & BLE Experience at Scale
The companion app frequently crashed for more than 10,000 daily active users, and BLE connectivity between the patches and the app was unstable across iOS and Android devices. Data sync often failed or stalled, interrupting HRV capture and real-time breathing guidance. Limited telemetry made it difficult to trace issues or understand failure patterns, which reduced adherence to breathing exercises and overall engagement.
Risky OTA Firmware Updates
OTA firmware updates often failed, occasionally bricked devices, and sometimes blocked clean rollbacks. There was no integrated, automated way to detect these issues before rollout. Every firmware push carried operational risk, put devices in the field at risk, and slowed the release cadence.
Complex & Inconsistent Sensor Data
Sensor data coming off the patches was complex and sometimes inconsistent, which made it difficult to translate raw readings into interpretable stress patterns. Without robust algorithms and visualizations, it was hard to deliver actionable insights and guided breathing experiences that felt intuitive to users.
Rising Support Costs & Slow Releases
The combination of crashes, connectivity issues, and firmware failures led to high support volumes and lengthy root-cause analysis. QA cycles were largely manual and fragmented across devices. This slowed down releases, reduced sprint predictability, and put pressure on the team’s ability to scale the product.
Approach & Solutions
1. Rebuild the Companion App for Stable, Secure HRV Sync
We redesigned the mobile experience and core sync stack to make BLE connectivity and data flows resilient across OS versions and device types. The focus was on reliability first, then performance and usability.
- Redesigned the BLE companion app and rewrote the sync stack with asynchronous retries and encrypted caching.
- Validated performance and connection stability across 25+ device and firmware combinations on iOS and Android.
2. Harden OTA Updates to reduce Risk in the Field
We treated firmware as a first-class release artifact and integrated it into a disciplined CI and regression flow. The goal was to detect OTA issues before rollout and reduce the chance of bricked devices or blocked rollbacks.
- Integrated a real-time regression lab with CI pipelines to simulate OTA scenarios and detect failures early.
- Standardized OTA test paths and validation gates so only proven firmware builds reached production devices.
3. Improve Traceability from Crash to Root Cause
We instrumented the platform so the team could see what was happening across app, BLE, and firmware layers in production environments. This reduced mean time to detect and resolve issues.
- Enhanced post-market surveillance with telemetry that captured crashes, connection failures, and firmware update health.
- Centralized logs and metrics to make it easier for engineers and QA to trace issues back to specific devices, OS versions, and firmware builds.
4. Turn Complex Sensor Data into Guided, Engaging Experiences
We focused on making sensor outputs interpretable and usable, so each session could deliver meaningful stress-management support rather than just raw readings.
- Developed algorithms for breathing sensor interpretation that translated HRV and breathing patterns into clear stress signals.
- Built interactive, gamified stress-management visualizations to guide users through real-time breathing exercises and reinforce adherence.
5. Make Quality Scalable across Devices, Sprints, & Releases
We implemented sustainable QA frameworks so the client could move faster without sacrificing reliability. Automation and device coverage were central to this approach.
- Implemented QA frameworks with automated regression testing and multi-device validation to cover critical user journeys and device/OS combinations.
- Integrated QA results into centralized dashboards to improve sprint predictability, release readiness visibility, and cross-team alignment.
Impact & Results
The client moved from a fragile, difficult-to-scale platform to a production-grade, clinically validated product that users and stakeholders could trust.
Connectivity stabilized, crashes were eliminated, firmware updates became safer, and QA cycles became faster and more predictable. This enabled the team to ship features and fixes more frequently, at lower operational risk and cost.
Connection success rate
The platform achieved a 97% connection success rate, up from 78%, across all OS and firmware combinations. Users experienced more reliable BLE pairing and HRV sync, which increased completion rates for breathing exercises and reduced session drop-offs.
Crash elimination
We resolved more than 30 major bugs and eliminated app crashes in the validated release set. This stabilized the daily experience for over 10,000 users and reduced support tickets tied to app failures.
Faster firmware validation
QA-firmware validation time dropped by 70%, from 5 days to 1.5 days. Faster, automated validation gave the team confidence to ship firmware updates more often without compromising safety.
Reduction in QA integration timelines
End-to-end QA integration timelines were reduced by 40%. Better automation, device coverage, and integrated regression flows allowed engineering and QA to move in lockstep rather than in long, sequential cycles.
Boost in sprint predictability
A centralized QA dashboard and more transparent quality metrics led to a 25% boost in sprint predictability. Product and engineering teams could plan releases with clearer expectations on risk and readiness.
Faster release cycle
The overall release cycle accelerated 6×, improving from more than 30 days per release to 5 days. This allowed the client to respond to user feedback, clinical insights, and market opportunities much faster, while keeping reliability high.
Ready to start the conversation?
Share your most pressing challenges and priorities with our team, and let’s explore what’s possible.