01
Connect
Talk to the equipment as it is.
Native drivers for PLCs, CNCs, robots, meters and OPC UA servers, plus devices that publish to the gateway’s built-in MQTT broker. No PLC program changes, no per-machine middleware.
Platform
One data layer, from the register to the decision.
Proxus collects from the equipment you already run, gives every signal a modelled address, governs who can see it, and delivers it wherever the work happens, without a separate integration per system.
- Collects from
- 66 protocols: PLCs, CNCs, robots, meters, OPC UA servers and devices that publish to the built-in MQTT broker
- Delivers to
- 18 destination systems: databases, message brokers and cloud services, and MES and ERP through them
- Runs on
- Your own servers, private cloud or air-gapped sites, Docker or Kubernetes
- Scales from
- From a single host to a multi-site fleet reporting into a central hub
How the data moves
Four steps, from the register to the system that uses it.
Proxus is not a driver pack and not a dashboard tool. It is the layer where industrial data is collected, given meaning, governed and delivered, so the same pipeline serves operations, IT and AI.
02
Model
Turn tags into assets.
Map raw registers to named signals with units, scaling and asset context, then publish them into a Unified Namespace that stays identical across every site.
03
Govern
Know what happened, and why.
Role-based access, an audit trail of configuration changes, edge buffering with store-and-forward, and operational logs per gateway.
04
Activate
Put the data to work.
Built-in historian, dashboards and reports for plant teams; streaming and database targets for IT; Copilot answers from live plant data and proposes changes you approve.
Working surfaces
The whole path, in one platform.
Connectivity, edge processing, the namespace, rules, functions, dashboards, delivery and Copilot. Explore the operating views and the architecture behind them.
Industrial connectivity
Device connectivity
Connect PLCs, CNCs, robots, meters and OPC UA servers, then define the tags you need and how often they are read.
Explore connectivityAt the line
Edge runtime
Collect, transform and buffer at the line. Evaluate rules, detect anomalies and run your own functions before the data moves.
Explore edge runtimeUnified Namespace
UNS Explorer
Browse every signal by enterprise, site, area, process cell, equipment module and device, with the live value and last update on every node.
Explore unified namespaceRules and alarms
Rule engine
Evaluate live values against conditions, raise alarms and trigger downstream actions without a separate SCADA layer.
Explore rules
Custom logic
Functions
Extend the pipeline with your own C# functions when a standard block is not enough.
Explore functionsConceptual architecture
Visualization
Dashboards and reports
Operational views and scheduled reports for plant and management teams, built on the same governed data.
Explore dashboardsIT and OT delivery
Integrations
Deliver modelled data to databases, message brokers, cloud services and business systems, per target and per strategy.
Explore integrationsAI Copilot
Proxus Copilot
Ask in plain words from any screen (Ctrl+K): why a device is silent, what changed, which alarms are open. Copilot reads the live namespace with your permissions and proposes rules and changes on a card. A change is applied only after you approve it, and the audit trail records it under your name.
Explore copilotCapabilities
What the platform actually does.
The parts teams configure first, with the equipment and the systems they already run.
Native drivers for the equipment on your floor.
Define a source, pick the protocol, list the registers you need and choose how often they are read. Proxus handles polling, reconnection and value change detection so equipment integration stops being a scripting project.
Read more- PLC
- Siemens S7 · Allen-Bradley · Mitsubishi · Omron · Beckhoff · Delta · LSIS
- CNC & robots
- Fanuc (FOCAS over Ethernet) · KUKA · Yaskawa · Haas · Heidenhain · MTConnect
- Standards
- OPC UA · Modbus TCP, plus RTU/ASCII over TCP · MQTT (built-in broker) · SNMP
- Browse
- OPC UA · MTConnect · SNMP · Allen-Bradley address spaces
Decide at the line, not three systems later.
Evaluate rules on live values, raise alarms, run statistical anomaly detection and extend behaviour with C# functions when a standard block is not enough. Results are buffered locally and forwarded when the link returns.
Read more- Rules
- Expression rules (thresholds and conditions) and anomaly rules, evaluated as each value arrives
- Anomaly detection
- Z-score, IQR, moving average, rate of change, drift, flatline and gap, on live streams
- Custom logic
- C# user-defined functions with the Proxus SDK
- Resilience
- Local buffering for the hub and every outbound target, written in order on reconnect
One address space every system agrees on.
Signals are published into an ISA-95-shaped hierarchy (enterprise, site, area, process cell, equipment module, device, tag) with units, scaling and metadata attached. New sites reuse the same model instead of inventing another naming scheme.
Read more- Structure
- Enterprise / site / area / process cell / equipment module / device / tag
- Context
- Units, scaling, data type and quality
- Reuse
- Device profiles shared by identical machines, on every site
- Access
- Subscribe by topic pattern from any authorised consumer
Serve operations and IT from the same pipeline.
Plant teams get dashboards, reports and alarms out of the box. IT gets the same governed data pushed into the systems they already run, with per-target transport strategies such as pass-through or on-change.
Read more- Databases
- PostgreSQL · SQL Server · Oracle · MySQL · Firebird · ClickHouse · InfluxDB · TDengine
- Streaming
- Kafka · RabbitMQ · MQTT · Sparkplug B · Redis
- Cloud & tools
- AWS IoT Core · Azure IoT Hub · Azure Data Lake Gen2 · REST endpoints · Splunk
- In-platform
- Historian, dashboards, scheduled reports and alarm workflows
Ask the plant in plain words; approve what changes.
Proxus Copilot is built into the hub and answers from live plant data with the asking user’s permissions. Changes come back as proposals you approve. Your own AI agents reach the same governed model through an MCP server, under the same roles and audit trail.
Read more- Copilot
- Built into the hub, Ctrl+K on any screen, answers with your permissions
- Changes
- Proposed on a card, applied only after approval, logged under your name
- AI service
- Claude, OpenAI, Azure OpenAI, Claude on Amazon Bedrock or Google Vertex AI, or a local Ollama model
- MCP
- For your own agents: signs in as a Proxus user, can be limited to reading, rate-limited, every call logged
Architecture
Built like plant infrastructure,
not like a web app.
One deployment model scales from a single host to a multi-site fleet reporting into a central hub, without changing how the data is modelled or who is allowed to see it.
- Edge first
- Collection, rules and buffering run next to the equipment, so the line keeps producing data when the network does not.
- Configured from the hub
- Gateways keep no configuration of their own: it comes from the hub. They authenticate, receive their devices, rules and functions and start working, so a failed unit is replaced, not rebuilt.
- Durable backbone
- Sites connect to a central hub over a durable, replayable message stream, with one topology from a single host to a multi-site fleet.
- Purpose-built storage
- Configuration and asset objects sit in a transactional store; telemetry and operational logs sit in a columnar store built for volume.
- Your perimeter
- On-premises, private cloud or air-gapped. Deployed with Docker or Kubernetes, operated by your own team.
Proven in production
Deployed across manufacturing, mining and energy.
-
Manufacturing
Digital factory data management
MEXT Digital Factory
Read the case study -
Mining & energy
Mining, fleet and solar on one platform
Kaltun Mining
Ask us about this deployment -
Energy
Grid monitoring at the edge
CK Enerji
Ask us about this deployment -
Manufacturing
Unified Namespace with Oracle ERP
Pilenpak
Ask us about this deployment
Your data, your infrastructure, your rules.
Deployment & security-
Inside your perimeter
On-premises, private cloud or air-gapped. Plant data does not have to leave your network.
-
Role-based access
Users, roles and object-level permissions across configuration, live data, screens and APIs.
-
Encrypted transport
TLS on every link: edge to hub, the hub’s own services, the built-in MQTT broker and outbound integrations, with certificates from the built-in CA or your corporate CA.
-
Auditable
Configuration changes are recorded in the audit trail, and pipelines keep revisions you can compare.
Next step
Point it at one of your lines.
The fastest way to judge a data layer is to watch it read your own equipment. Bring one machine and one destination system, and we will build the path in front of you.
- 01
Scoping call
Equipment, protocols, sites and the systems the data has to reach.
- 02
Technical walkthrough
A working demo against a scenario that resembles your plant, not a generic tour.
- 03
Pilot plan
A defined first line or site, with the deployment and security questions answered up front.
- 04
Roll-out
The same templates bring the next lines and sites online.