Built by CodesClue

AirShield

Real-time air quality intelligence and compliance reporting for hospitals and clinics.

Industry
Healthcare and IoT
What we did
IoT platform and companion apps
Platforms
iOS, Android, Cloud

Results

Active Connected IoT Devices
50+
Real-Time Telemetry Accuracy
98%
Operational Efficiency Boost
60%
Reduction in Manual Effort
70%

Overview

AI-powered IoT companion platform delivering real-time air quality intelligence, predictive maintenance, and compliance reporting for healthcare and regulated indoor environments.

Smart air protection for critical indoor spaces

AirShield connects to proprietary air purification hardware across regulated indoor environments, providing 24/7 real-time sensor telemetry, automated maintenance alerts, and audit-ready compliance exports all through native iOS and Android companion apps backed by a 100% serverless AWS infrastructure.

The platform serves hospitals, clinics, assisted living facilities, schools, and commercial spaces where air quality directly impacts patient safety, regulatory standing, and operational continuity. Built in collaboration with the US Department of Defense, every technical decision was made to defense-grade standards of device reliability, data integrity, and environmental performance.

AirShield product screen 1
AirShield product screen 2

Our approach

We engineered a scalable IoT platform that turns real-time air quality data into actionable insights for healthcare operators. From telemetry and AI-based maintenance to audit-ready reporting and simplified UX, the solution was built to improve reliability, compliance, and day-to-day operational efficiency.

  • Real-Time Telemetry Architecture

    Designed a continuous IoT data pipeline using AWS IoT Core with bidirectional MQTT communication, capturing pathogen levels, PM2.5, disinfection cycle data, and device performance metrics around the clock. Versioned DynamoDB schemas accommodate new sensor types as hardware evolves without disrupting existing telemetry streams. Offline buffering was built into the companion apps to handle healthcare network restrictions with zero data loss.

  • AI-Driven Preventative Maintenance

    Integrated a machine learning layer that continuously analyses device performance trends to detect early degradation signals before failures occur. Purification cycles adapt automatically based on detected air quality patterns, reducing unplanned downtime in live healthcare environments. Maintenance advisories are surfaced directly to facility operators through push notifications, enabling action before critical thresholds are breached.

  • Compliance and Audit-Ready Reporting

    Built exportable air quality and disinfection cycle reports structured for regulatory submissions and accreditation audits across all facility types. Full audit trail maintained with historical retention covering device uptime, firmware update events, zero-emission compliance status, and timestamped telemetry records. Role-based access ensures compliance viewers get read-only access to reports without exposing operational controls.

  • UX for Non-Technical Operators

    Post-MVP UX iteration was driven directly by real-world feedback from deployed facility staff with no technical background. Complex air quality metrics were translated into clear, actionable dashboards with threshold alerts, device status indicators, and maintenance advisories designed for maintenance operators not engineers. Significant iteration cycles refined the interface based on how facility teams actually used the product in the field.

AirShield product screen 3
AirShield product screen 4
AirShield product screen 5

Challenges and how we solved them

Healthcare IoT is a uniquely demanding environment. Every technical decision carried real consequences for patient safety, regulatory standing, and operational continuity.

Each challenge was met with an engineering decision that protected uptime, data and the people relying on the air quality readings.

  1. Challenge

    High-Frequency Telemetry at Scale

    Continuous sensor data from many simultaneous devices across multiple facilities threatened write bottlenecks at the data layer. The challenge was sustaining reliable ingestion at high throughput without introducing latency or data loss across an expanding device fleet.

    Solution

    Batched Writes and DynamoDB Optimization

    Addressed telemetry throughput through batched write patterns and DynamoDB table configuration optimized for high-frequency IoT workloads. This approach sustained reliable ingestion across growing device fleets without traditional backend scaling overhead, leveraging AWS managed services to handle volume automatically.

  2. Challenge

    UX for Non-Technical Operators

    Complex air quality metrics PM2.5 readings, pathogen detection levels, disinfection cycle data, needed to be understood and acted upon by facility maintenance staff with no data or engineering background. The risk was building a technically accurate product that operators couldn't effectively use in the field.

    Solution

    Post-MVP UX Iteration from Real Operators

    Conducted structured UX iteration cycles based on direct feedback from facility managers and maintenance staff using the deployed product. Dashboard layouts, alert thresholds, and notification language were refined specifically for non-technical users simplifying complex air quality data into clear status indicators and actionable advisories.

  3. Challenge

    Healthcare Network Restrictions

    Strict firewall policies in hospitals and clinics meant that network connectivity gaps were unavoidable during day-to-day operations. Any architecture that assumed continuous connectivity would result in telemetry data loss at precisely the environments where data integrity mattered most.

    Solution

    Robust Offline Buffering

    Built local data buffering into the iOS and Android companion apps to capture and queue telemetry during network-restricted periods. Data syncs automatically when connectivity is restored, ensuring zero telemetry loss regardless of hospital or clinic firewall conditions.Designed the platform to be flexible, allowing quick adaptation to changing legal requirements across different regions.

  4. Challenge

    Live-Environment Firmware Updates

    Updating firmware across active healthcare deployments without interrupting purification cycles was a uniquely high-stakes challenge. A failed or poorly sequenced OTA update in an operating theatre or ICU carries direct consequences for patient safety and facility compliance status.

    Solution

    Staged OTA Firmware Rollouts with Rollback

    Designed an OTA update mechanism with staged rollout sequencing that protects active purification cycles throughout the update process. Incremental deployment ensures only a subset of devices update at any time, with rollback capability ready if anomalies are detected maintaining operational continuity across live healthcare environments.

AirShield product screen 6

How it works

AirShield begins at the device level IoT hardware running C/C++ firmware captures continuous sensor readings including PM2.5 particulate levels, pathogen detection, and disinfection cycle status. Data transmits in real time over AWS IoT Core via bidirectional MQTT, with offline buffering active on the companion app side to handle any periods of restricted network access common in healthcare environments.

  1. Facility managers and operators access the platform through native iOS or Android companion apps secured with AWS Cognito and Microsoft MFA. The live dashboard surfaces air quality readings, device status, and threshold breach alerts via push notification presented in clear, non-technical language designed for maintenance staff rather than engineers. Alerts trigger automatically when sensor readings cross configured thresholds, enabling rapid response before air quality issues escalate.

  2. On the data side, telemetry flows into DynamoDB with versioned schemas that accommodate new sensor types as hardware evolves. The AI and ML layer runs continuously in the background, analyzing device performance trends to detect early degradation signals and adapting purification cycles based on detected air quality patterns reducing downtime before it starts.

  3. Multi-tenant facility management keeps each deployment fully isolated at the AWS infrastructure layer. Hospitals, clinics, assisted living facilities, schools, and commercial spaces each operate within their own scoped environment, with role-based access separating Facility Administrators, Operations Staff, and Read-Only Compliance Viewers. Compliance reports covering air quality readings, disinfection cycle history, zero-emission status, and full audit trail are exportable on demand for regulatory submissions and accreditation audits.

  4. Firmware updates roll out via a staged OTA mechanism designed to protect live purification cycles, with rollback capability ensuring healthcare environment deployments are never disrupted by an update sequence.

All these allowed me to really interface with Ketan and his team quite closely and understand how excellent they are at communications, time management and delivering work.

Read the transcript

Hi, I'm Stefan Levine. I'm a real estate professional in Jupiter, Florida in the United States. I have a background in telecom and technology, and over the years I've worked closely with Ketan and his team on a number of projects. I ended up advising a real estate client on a rebuild of a dated code set for some crucial software to their company's hiring process, as well as developing an app for intercompany use for some of their core processes. All these allowed me to really interface with Ketan and his team quite closely and understand how excellent they are at communications, time management and delivering work. I remain in contact regularly with Ketan primarily, and even when there aren't projects in discussion, as iron sharpens iron, the CodesClue team allows me to remain with them at the forefront of software application and AI development, and keep track of what's going on in the world. So I can't emphasize enough how worthwhile it is for you to speak to them, understand the questions that they ask and the way they respond to your requirements, and how that's going to have such a positive impact on your jobs. Thank you.

Stefan LevineReal estate professional, Jupiter, Florida

Have a complex IoT or healthcare platform to build?

From embedded firmware through to cloud infrastructure and companion apps CodesClue delivers end-to-end. Let's talk about your product.

  1. Answer two quick questions and tell us about the project.
  2. We read it and reply with questions or a plan.
  3. You get a written estimate with milestones before anything is signed.
Or book a 30-min call (opens in a new tab)

    Build together with CodesClue

    Next project

    Subjects Master

    Game-based study guides for teachers, schools and homeschool families: we took over the live platform, made it fast, and keep improving it.

    More work like this

    All 17 case studies

    HealthcareAI product build

    Therapix.AI

    AI clinical documentation that turns therapy sessions into structured notes and SOAP reports.

    62% reduction in documentation time

    • React.js
    • Node.js
    • PostgreSQL
    • +3 more

    Healthcare, AI SaaSDesigned and built by CodesClue

    Therapix.AI UI/UX design

    Design that speaks the clinician's language.

    Steel manufacturingGate check-in app with Emirates ID scanning

    ArcelorMittal

    A mobile app for gate and security staff: scan a visitor's Emirates ID, check them in, and check them out against the right employee and vendor.

    • Flutter
    • Node.js