AI-Powered Connected Device Management for Diagnostics, Analytics, and In-Clinic Guidance

The client is a global med tech company in professional skincare. Its provider network includes over 30,000 estheticians across 90+ countries. As a US-based FDA-registered manufacturer of equipment and consumables, the client supports clinics with patented device technology and treatment serums.

Device reliability and service continuity were becoming difficult for the client to manage at scale. Providers faced interruptions from issues such as low flow, clogs, and noise disruptions. These events created downtime and inconsistent treatment experiences.

Support and operations teams lacked a unified, real-time view into device health, usage, and consumables. Inventory tracking was not connected to actual device consumption, which contributed to shortages and delays in treatment.

As device adoption expanded across locations, operational blind spots also began to appear in day-to-day use.

• Maintenance steps were not completed consistently, which increased the risk of performance drops and failures.

• Software upgrades were delayed across devices, which created version drift and risk of downtime.

• Devices were moved between locations without authorization, which disrupted tracking and planning.

• Expired consumables were occasionally used, which affected service quality and created a safety risk.

• Usage trends were hard to track across time and location, limiting staffing and sales planning.

Given the nature of clinical-grade devices, the platform needed robust security controls and access governance, without slowing down workflows.

Coditas designed and delivered a connected device ecosystem that combined IoT telemetry, remote diagnostics, an AWS-based data lake for analytics, and an embedded GenAI assistant to support providers during treatment.

Remote Monitoring and Digital Twins

IoT sensors streamed device telemetry into a cloud platform, creating digital representations of each device. This gave support teams continuous visibility into health signals and operational status.

Rules, Alerts, and Secure Connectivity

The platform enabled configurable rules for event detection and notifications. Secure remote connectivity supported diagnostics and remote actions while protecting sensitive operational data.

Role-Based Access and Audit-Ready Controls

User management was built to support provider and internal support roles. Access to device data and actions was controlled through permissioned workflows suited for regulated environments.

Data Engineering Foundation on AWS

We implemented a cloud native data lake and streaming pipeline to ingest, curate, and query IoT telemetry at scale. This data foundation supported monitoring, reporting, and provider-facing product features that rely on historical and real-time session context.

• Set up real-time telemetry ingestion using AWS IoT Core and Amazon Kinesis Streams

• Delivered raw device data into Amazon S3 through Amazon Kinesis Firehose, batching records to reduce small file overhead

• Standardized payloads by flattening JSON with AWS Lambda into a consistent schema before storage

• Partitioned the S3 data lake by year, month, day, and hour to speed up downstream queries

• Registered datasets in AWS Glue Data Catalog to manage table definitions and partition metadata

• Enabled analysis through AWS Athena and routed outputs into R and Python for reporting workflows

• Automated partition refresh using MSCK REPAIR TABLE to keep fresh data queryable without manual updates

• Curated datasets to support maintenance reporting, location-based performance analysis, usage trends, and session-level context retrieval

Operational Analytics for Providers and Support Teams

An analytics dashboard surfaced real-time device health, usage trends, and operational patterns. Teams could spot recurring issues, prioritize servicing, and plan capacity based on usage behavior.

GenAI Assistant on the Device Experience

The GenAI assistant supported providers across the treatment journey. It pulled past treatment history when available, supported a real-time skin assessment using the device camera, and suggested a personalized treatment plan based on skin conditions, history, and preferences.

During treatment, it displayed progress, captured a session summary, and logged provider-consumer interactions. After treatment, it suggested relevant products and supported a purchase flow. Provider login unlocked tools to review consumer history, monitor inventory signals, and trigger replenishment through QR code scans.

When troubleshooting was required, the assistant guided self-diagnostics. It also created support tickets and shared status updates when issues needed human escalation, while collecting feedback to improve future sessions.

The engagement improved device uptime, reduced operational friction, and strengthened the provider experience through connected workflows and reliable data.

• 30% revenue increase from the client’s IoT-enabled device line

• More than 25% increase in supply sales through better usage visibility and replenishment triggers

• Over 60% reduction in field service coverage through remote diagnostics and device management

• Device downtime reduced by 20% through better monitoring and faster intervention on maintenance issues

• Operational costs reduced by 25% through a scalable, cloud native data platform

• Provider satisfaction improved from 3.5 to 4.8 on a 5-point scale, based on a provider survey

This engagement shows the value of pairing connected device telemetry with a scalable data foundation and workflow-led experiences for providers. Coditas helped the client move from reactive service cycles to proactive monitoring, faster resolution, and data-backed operational decisions.

Want a GenAI-backed connected device ecosystem that runs reliably in production? Coditas can help.

Get in touch with our experts to discuss a customized solution.