We are building a cloud-connected monitoring platform for industrial power generation equipment. The core hardware is a Linux-based industrial IoT gateway with a built-in CAN interface and cellular modem. The equipment controller exposes operating data over a standard J1939 CAN interface.

You will receive a development unit and work through the full stack: reading CAN data off the controller, building a robust edge application, and connecting it to a cloud-based data pipeline. This is a greenfield project with real hardware from day one. Future scope includes a customer-facing app for remote monitoring and control.

What You Will Build

• A CAN reader service on Debian Linux using SocketCAN and python-can, decoding J1939 PGNs from the equipment controller
• Filtering, aggregation, and store-and-forward buffering logic on the edge device
• An MQTT publisher with telemetry and event-based transmission modes, connecting to a cloud-backed broker over cellular
• Docker containers for clean deployment and OTA update delivery
• Contribution to the cloud pipeline (ingestion, storage, and alerting)

Requirements

• Linux systems programming -- writing systemd services, managing processes, working at the terminal on ARM Debian
• Python -- async patterns, socket/serial communication, background service design
• CAN bus -- hands-on experience with SocketCAN, candump, cansend, or python-can on real hardware
• J1939 -- understanding of PGNs, SPNs, and DM1 fault messages
• MQTT -- pub/sub, QoS, TLS, broker interaction
• Docker -- building and deploying containers on ARM Linux
  

Nice to Have

• Experience with industrial generator controllers or similar heavy equipment controllers
• Cloud data pipeline experience (ingestion, warehousing, serverless functions, or stream processing)
• OTA update systems for embedded Linux
• Data visualization tools (BI dashboards or time-series)
• Fleet-scale IoT -- SIM management, device provisioning pipelines, certificate management
• Mobile or web app development