Industrial operations platform

Public beta

Build, simulate, and operate heavy industry.

One platform from the PLC to the cloud. Author equipment from a typed standard library, prove it in closed-loop simulation, and operate it through control screens your team assembles in real time — with AI that explains the plant, never controls it.

Request a demo Start free

Built on 20+ years of industrial automation delivered to Tier-1 Australian mines.

Not a whiteboard. A platform built on production PLC code from real plants.

20+years in automation

112UDTs of production code

57AOIs of production code

1vertical-agnostic platform

One architecture, three legs

The full lifecycle of an industrial process

Vertical-agnostic by design — its reach is bounded only by standard-library coverage. Mining is the first library, not the ceiling.

Build

Author device types, processes, and applications against a typed industrial standard library — templates that mirror Rockwell AOIs and Ignition UDTs, extended with UI, alarms, and AI context.

Simulate

Run closed-loop simulation that behaves like the real PLC — native sim or a full Phoenix Contact vPLCnext in a container. Train operators, run what-if scenarios, and validate logic before it touches the field.

Operate

Server-rendered control screens assembled in real time from operator queries and live plant state — streamed to the browser over SignalR. A renderer-agnostic semantic UI tree drives the browser and the csy CLI from the same server — one tree, two live renderers, native next.

Capabilities

What's running today

The platform that took a dam pump station from PLC to cloud — live across the full chain.

Server-driven UI

Control screens rendered in real time over SignalR — no SPA to maintain. A renderer-agnostic semantic UI tree lets one server-side source paint the browser and the csy CLI — rolling out surface by surface.

Closed-loop simulation

A native Sim runtime and a containerised Phoenix Contact vPLCnext run the same control logic as a physical PLC — for what-if, training, and commissioning.

Device type system

Reusable equipment templates with tags, alarms, commands, UI faceplates, and simulation profiles — a shipped standard library distilled from 112 UDTs and 57 AOIs of production control code.

IO AI code generation

Deterministic generation of the PLCnext control program from device-type mappings — same input, same PLC code, every time.

Provenance historian

Every sample carries its source — measured or simulated today, with imported, inferred, replayed, and workspace classes built into the same schema. Auditing measured-versus-derived becomes a query, not tribal knowledge.

Living electrical model

A connected model of site power distribution that computes voltage drop, fault levels, and load — and redraws the single-line diagram when the model changes.

Offline-tolerant edge

Edge runtime on PLCnext edge controllers with store-and-forward, so the site keeps reporting when the link to the cloud drops. On-device ML inference for anomaly detection is on the roadmap.

Hard multi-tenant isolation

Subdomain-per-tenant with separately audited isolation layers and a hostile cross-tenant probe suite. Your platform data stays in Azure Australia East.

Explore the platform

Safety first

AI explains what the PLC is doing.
It never takes control.

Every control action goes through the PLC on the established deterministic path. The AI — cloud advisory reasoning today, hosted in-region, with site-local models and edge ML on the roadmap — is advisory and diagnostic: it recommends and explains anomalies, but it cannot command equipment.

Hard interlocks stay in IO modules and safety relays, consistent with IEC 61511 and ISO 13849. That line is non-negotiable for mining safety culture — and it's built into the architecture.

Australian data residency

All platform data in Azure Australia East — and AI inference stays onshore, in Australian regions only.

Air-gap ready

The edge keeps running offline at the substation; the AI tier can be disabled entirely for strict sites. On-device ML and self-hosted models are on the roadmap.

One codebase, two artifacts

The cloud service and the on-prem site server are published from one codebase — cloud is a topology, not a separate product. Download the site server and csy CLI →

See it run on your process.

A 30-minute walkthrough on a live plant model — pumps, conveyors, MCC sections — running in our simulation engine.

Request a demo Browse the library