Monitoring a distributed energy installation reliably means designing for remote, intermittently-connected hardware from the start, not adapting a connectivity assumption that only holds in the lab. Rivix builds the sensing, firmware and telemetry pipeline as one system.
Distributed thermal and electrical sensing hardware built for installations nobody visits daily.
Explore hardware →Real-time diagnostics on automotive- and industrial-grade MCUs, plus protocol-specific gateway integrations.
Explore firmware →Fleet-wide dashboards giving visibility into performance and faults across distributed installations.
Explore cloud →Web dashboards built for operators watching a fleet, not a single device.
Explore mobile & web →Fault detection at fleet scale — on-device or cloud, chosen by the constraint.
Explore AI & Intelligence →The design assumption is that an attacker can get physical access to the device: JTAG/SWD debug ports disabled or locked down in production builds, key material stored in hardware-backed secure elements rather than plain flash, and tamper-evident or tamper-responsive enclosures. Firmware runs signed and verified at every boot, extended to signed, authenticated OTA updates with anti-rollback protection.
A grid-edge device's telemetry, config, and OTA path is kept as a separate trust boundary from utility control systems — defense-in-depth rather than a single perimeter.
RF mesh (hop-to-hop to a collector) suits dense residential/urban deployments. Cellular — LTE-M, NB-IoT, or standard LTE — works per-device with no collector infrastructure but carries a recurring data-plan cost, better suited to sparse or rural areas. Power line communication rides the existing distribution wiring, with signal quality sensitive to the grid segment. Utility rollouts often combine mesh or PLC in dense areas with a cellular tail for remote ones. Reporting at 15- or 5-minute granularity instead of monthly/daily totals enables demand response and outage detection, at the cost of data volume at fleet scale.
The battery management system is treated as the authoritative source for cell-level health and safety; site monitoring aggregates BMS, inverter and production data rather than duplicating BMS function. Time-sensitive protective actions — over-voltage charge cutoff, fault alarms — run locally at the edge, independent of connectivity. Our EV thermal-monitoring module runs cell-level sensing into an automotive-grade MCU, with diagnostics evaluated locally and reported over the in-vehicle bus rather than depending on an external network.
Pulled from our case-study library — same AI/human split shown in full, industry context included.
A distributed monitoring module captures thermal conditions across an EV battery pack and identifies abnormal temperature behavior for battery-management and diagnostic systems.
A connected platform collects operational telemetry from distributed solar micro-inverters and provides fleet-wide visibility into performance, device health and faults.
Designing for remote, intermittently-connected hardware from the start, not adapting a lab-only assumption.
Read the article →Physically exposed grid-edge hardware needs secure boot, signed updates, and segmentation as a baseline.
Read the article →How often a meter reports, and over what network, sets the ceiling on cost, battery life, and utility value.
Read the article →Grid-edge security is a genuinely different problem from a consumer gadget — it's designed as a baseline, not an add-on, from the first schematic.
FreeRTOS, Zephyr and bare-metal on the RTOS side; BLE, Wi-Fi, LoRa, Zigbee and cellular on connectivity, chosen based on your power and range constraints, not by default.
Both, chosen by the constraint — real-time and privacy-sensitive workloads run on-device; fleet-wide pattern detection runs in the cloud.
Yes — BOM optimization and sourcing (including lifecycle/obsolescence risk) is part of the hardware engagement, not a separate line item.