Telematics Platform Management

Every machine carries a persistent embedded Linux logger that captures its CAN networks around the clock. Loggers connect over cellular or satellite through a WireGuard VPN to an AWS-hosted backend, which serves a fleet portal and Grafana dashboards to engineers and customers.

The fleet spans Canada, the US and South America, on everything from oilfield rigs to underground drills and haul trucks. I took ownership of the platform in late 2025 and have run it since.

• Root-caused intermittent data loss to CAN frames with invalid timestamps, then filtered them at the source.
• Fixed a date-rollover bug at midnight UTC and rolled the fix out to the fleet over the air.
• Fixed log compression timing out on dense 500 kbps networks, and added CAN error frames to the logs.
• Recovered loggers through the bootloader, corrected SD card partitioning, and wrote the flashing procedures.

• Uptime tracking, GPS appended to logs, on-demand manual recordings and live cellular diagnostics.
• A local dashboard on every device, reachable remotely through HTTPS forwarding on the cloud host without joining the VPN.
• Status emails to the service team for preventative maintenance.
• Machine hour tracking end to end: new hour meters in the vehicle controller's Simulink model with non-volatile storage, broadcast over CAN, read by a Linux service on the logger and pushed to a customer's maintenance platform API.

• Brought the portal back after a critical remote-code-execution vulnerability (CVE-2025-55182), then hardened the deployment with process persistence, dependency audits and tighter security policy.
• Role-based access in Grafana so customers only see their own machines.
• Wrote the platform roadmap with AWS cost analysis, and evaluated AWS IoT, MQTT and self-hosted VPN key management as next steps.
• Multi-IMSI SIM provisioning for South American sites, and haul-truck installs at an operating mine, including the wiring harness design.