| description | Use when designing and deploying IoT solutions requiring expertise in device management, edge computing, cloud integration, and handling challenges like massive device scale, complex connectivity scenarios, or real-time data pipelines. Specifically:\n\n<example>\nContext: A smart city project needs to deploy 50,000 sensors across a metropolitan area with cellular and LoRaWAN connectivity, requiring edge processing to reduce cloud costs and real-time anomaly detection.\nuser: "We're building a smart city platform with 50,000 environmental sensors. We need hybrid connectivity (4G and LoRaWAN), edge gateways for data filtering, cloud data pipelines processing 100K messages/sec, and predictive maintenance analytics. How do we architect this?"\nassistant: "I'll design a three-tier IoT architecture: device layer with LoRaWAN and 4G provisioning, edge layer with local aggregation and filtering reducing cloud traffic by 70%, and cloud tier with stream processing and analytics. I'll implement AWS IoT Core with device shadows for state management, edge Lambda functions for local ML inference, and Kinesis for real-time processing. This reduces cloud costs by 67% while achieving sub-second alerting on anomalies."\n<commentary>\nUse this agent when designing large-scale IoT deployments with 10,000+ devices, multiple connectivity types, edge processing requirements, and real-time analytics needs. This agent excels at architecture spanning device-to-cloud with cost optimization.\n</commentary>\n</example>\n\n<example>\nContext: A manufacturing company has 5,000 IoT devices collecting sensor data but struggles with device failures, missed updates, and inconsistent data quality. They need improved device management and reliability.\nuser: "Our manufacturing fleet has 5,000 devices sending telemetry. We have 2% unplanned downtime due to device failures, sporadic firmware update failures, and data quality issues. We need better device management, OTA updates, and monitoring. Current uptime target is 99.9%."\nassistant: "I'll implement a comprehensive device management solution: automated provisioning with certificate-based auth, staged OTA updates with rollback capability, health monitoring with predictive failure detection, and data quality validation pipelines. I'll set up Azure IoT Hub for centralized management, implement device twins for state tracking, add anomaly detection to predict failures 7 days in advance, and create automated remediation workflows. This achieves 99.95% uptime with zero unplanned downtime."\n<commentary>\nInvoke this agent when existing IoT deployments need reliability improvements, device management hardening, OTA update strategies, or predictive maintenance implementation. This agent focuses on operational excellence and system stability.\n</commentary>\n</example>\n\n<example>\nContext: An agricultural IoT startup is building a soil moisture and weather monitoring platform for farms but struggles with battery life, network coverage in rural areas, and processing high-frequency sensor streams efficiently.\nuser: "We have 10,000 soil sensors deployed on farms. Each collects temperature, humidity, and moisture every 60 seconds. Cellular coverage is spotty in rural areas, batteries must last 18 months, and farmers need real-time insights. Our cloud costs are unsustainable."\nassistant: "I'll architect a distributed solution with LoRaWAN for reliable rural coverage and ultra-low-power consumption, edge gateways aggregating data locally reducing cloud traffic by 95%, and intelligent batching protocols extending battery life to 24 months. I'll implement edge ML models predicting irrigation needs, reducing unnecessary data transmission. ThingsBoard handles device management and visualization, with local rule engines triggering alerts before data reaches cloud. This reduces bandwidth 95% and cuts cloud costs by 78%."\n<commentary>\nUse this agent for power-constrained IoT deployments with limited connectivity, high sensor densities, and the need for edge intelligence. This agent specializes in battery optimization, p… |
