The same building blocks behind a city-scale water grid and a four-building HVAC fleet. We draw the cloud / on-prem boundary based on data sensitivity and operating profile — not vendor preference.
One repeatable shape behind every deployment. We draw the cloud/on-prem line based on data sensitivity and the operating profile — not vendor preference.
Pick the right radio for the deployment's range, power, and density profile. We deploy across all of these and bridge them into the same upstream model.
| Protocol | Range | Power profile | Best fit |
|---|---|---|---|
| LoRaWAN | 2–15 km | Multi-year battery | Outdoor utility, agriculture, asset tracking |
| NB-IoT / LTE-M | Cellular coverage | Months–years | Structural, fleet, geographically scattered |
| Bluetooth LE / Mesh | 10–100 m | Months | Indoor occupancy, access, beaconing |
| Sub-GHz RF (433/915 MHz) | 0.5–5 km | Years | Dense mesh on private spectrum |
| Wi-Fi / 802.11 | ~100 m | Wall-powered | Camera, gateway, on-site dense data |
| MQTT / CoAP | — | — | Application-layer transport over any of the above |
The things you actually do with the platform once it's running — push config, send commands, pull data out, bridge a 3rd-party network.
Upload a single config file that defines which endpoints your hardware exposes for I/O — gateway, sensor, or controller. Auto-applies on the next handshake. Works cleanly with openWRT and similar open-source firmwares.
Send requests to devices, get status updates, create event schedules, change device configs remotely, run scripts, and send custom commands — all through the same API surface and audited per-tenant.
Download device status history as CSV / JSON, hit the time-series endpoints directly, or subscribe to the WebSocket stream. No support ticket required.
An on-prem instance can incorporate 3rd-party APIs into the same dashboard when the target endpoint sits on a parallel network — no exposing the secured network to the public internet.
Time- and location-triggered actions on device fleets. Schedule the watering run, fire a webhook when a fleet vehicle leaves the depot, or rotate device certificates on a cadence.
Tenant isolation at the data, API, and dashboard layers. White-label UI sits on top of the same core — same multi-tenant guarantees, your brand on the chrome.
A working library of device profiles and user-shared scripts for common hardware — browse what's already wired up, or contribute your own once you're deployed. Speeds up bring-up on anything that's been integrated before.
Stateless services per concern, observable from the gateway down, packaged for both managed cloud and customer-managed clusters.
Stateless ingest, fan-out, and analytics services that scale per-tenant. Packaged as containers for both Kubernetes and Fargate-style targets.
One authenticated edge for every device, partner, and dashboard. Policy, rate-limiting, and distributed tracing built in.
MQTT, Kafka-style topics, and WebSocket fan-out so applications subscribe instead of polling. Idempotent producers across the bus.
Time-series at the hot edge for sub-second queries; columnar warehouse for analytics and BI connectors for the reporting layer.
Sub-second control loops with deduplication and exactly-once semantics. For command/ack, geofencing, and SLA-bound alerting.
Drop our gateway in front of an existing micro-network, fleet system, or internal ERP. Keep the upstream data model you already have.
Cellular deployments don't end at the modem — they start there. We integrate with the standard connectivity-delivery platforms and stay aware of the call flow underneath, so issues get triaged against the right layer the first time.
| Layer | Coverage |
|---|---|
| Radio & modules | LTE · LTE-M (Cat-M1) · NB-IoT · Private LTE · 5G NR / mmWave-aware integration |
| Core elements | MME · S-GW / P-GW · PCRF · Diameter / RADIUS · HSS · IMS · SMS gateways |
| SIM & connectivity | Cisco IoT Control Center (Jasper) · Nokia WING · Ericsson DCP · multi-IMSI / eSIM provisioning · SIM-OTA |
| Operations | Triage runbooks · LTE call-flow tracing · device-behavior correlation · roaming & intercarrier escalation |
For enterprise customers and resellers, the platform isn't useful until it reports against the network's actual key performance and key service indicators. Built for monthly / quarterly business reviews.
A view of how the IoT service is performing per enterprise customer and per reseller — uptime, latency, error budget, roaming events, SIM lifecycle. Built for monthly / quarterly business-review cadence.
Proactive communication of planned changes; structured incident updates and root-cause analysis for unplanned events. Negotiated change windows with customer technical leadership.
ELK / OpenSearch for logs, Prometheus-style metrics, distributed tracing across the microservice fabric. Nagios- and PagerDuty-class alerting tied to SLO budgets, not raw thresholds.
Documented troubleshooting procedures route issues to the right resolver — Service Ops, Engineering, IT, Intercarrier, or device OEM — without playing email tennis.
Alerts are tied to the business outcome (commands acknowledged, packets ingested, dashboards available) rather than to component liveness.
A late-binding data warehouse that lets analytics evolve without forcing schema rewrites every time a new sensor family lands. Connectors for BI and reporting.
Don't confuse more data with more insight. Many solid findings are interesting but don't move the operational needle.
Don't confuse insight with value. Insight only matters when it changes a decision someone is already making.
Don't overestimate the ability to interpret the data. Even the best dataset gives only partial visibility into the real-world process behind it.
Don't underestimate the challenge of implementation. Leveraging telemetry requires the receiving organization to embrace new methodologies — a real time and capital investment.
Same platform, three deployment profiles. The decision is data sensitivity and ops capacity, not features.
| Target | Best when | Ops responsibility | Typical scale |
|---|---|---|---|
| Managed cloud | Public-facing data, fast spin-up, growing footprint | We run it end-to-end; SLAs apply | 10 – 100k devices |
| Customer cloud (BYOC) | Existing AWS / Azure / GCP investment, compliance baseline already established | You own the cloud account; we operate the workload | 1k – 1M devices |
| On-prem / air-gapped | Security-sensitive, no outbound, regulated industries | You operate; we hand over a versioned cluster bundle | Any |
The demo platform is a working multi-tenant build — log in or browse the read-only metrics view. If a project sounds close to one of ours, we'll walk through the reference architecture.