Smart Home IoT AI Monitoring

I’m spinning up a UPT-based initiative that needs hands-on IoT and artificial-intelligence engineering talent. The first milestone centres on a smart-home environment where multiple sensors and devices stream data that must be captured, analysed and displayed in real time. I already have basic hardware prototypes on the bench; what I need now is an end-to-end solution that reliably moves data from the edge to the cloud, applies lightweight AI for instant insights, and returns actionable notifications with minimal latency.

You should be comfortable designing embedded firmware (ESP32/Arduino-class boards or similar), setting up secure over-the-air updates, and wiring the data pipeline—MQTT or another broker, a fast time-series database, and a responsive dashboard (Grafana, React, or your preferred framework). On the AI side, expect to build or integrate models that can spot anomalies or usage patterns on the fly; Python, TensorFlow Lite, or Edge Impulse are all fair game as long as inference happens quickly.

Acceptance criteria for this phase:
• Architecture diagram covering device, gateway, cloud and mobile layers
• Working firmware that pushes sensor data every second with