Remote terminals commissioned once and then expected to run unattended — on a path with permanent latency, scarce space segment, weather in the topology, and no second route to fail over to.
What I do
Day-to-day operation of an iDirect hub and its remotes: link budgets, remote commissioning, carrier and modem configuration, and hybrid failover testing against terrestrial paths — plus the Cisco, Fortinet and MikroTik estate behind it, and the monitoring that flags a degrading terminal before the site calls.
The result
Terminals that stay up unattended, and a failover path that has been tested rather than documented. Designing under a hard bandwidth and latency ceiling changes how you build everything afterwards.