Platform, not projects
- Public copy
- Every deployment should strengthen reusable capabilities instead of creating a customer-specific fork.
ABOUT MERIKH
Building the industrial intelligence layer factories can rely on.
Merikh exists to help factories connect what is happening in the physical world with the people, systems, and decisions responsible for improving it.
Why Merikh exists.
Factories already generate valuable operational knowledge. Some of it lives in machines and control systems. Some lives in ERP, maintenance, quality, or production tools. Much of it remains in spreadsheets, messages, and the experience of people on the floor.
When those sources stay fragmented, teams spend time reconciling what happened instead of deciding what to do next. Merikh is being built to create a shared operational model—one that connects physical signals, production context, human knowledge, and accountable action.
Our vision: from Factory OS to Industrial Intelligence—and toward Physical AI.
The near-term job is to make factory reality connected, understandable, and actionable. Over time, that foundation can support role-based AI agents that reason with evidence, coordinate workflows, and assist with physical operations within clearly defined authority. The path to Physical AI begins with reliable data, operational understanding, and trust—not automation for its own sake.
The principles behind the platform.
Reliable data before AI
- Public copy
- Accuracy, context, and traceability come before advanced analytics or AI recommendations.
Understanding before automation
- Public copy
- The platform must understand assets, processes, constraints, and authority before it acts.
Human + AI collaboration
- Public copy
- AI should assist, recommend, and execute only with the permission and scope appropriate to the role.
Configuration over customization
- Public copy
- One reusable platform should adapt to factory reality without becoming a collection of separate products.
Every insight should create action
- Public copy
- The operating loop should move from data to insight, recommendation, ownership, action, and outcome.
Physical-world responsibility
- Public copy
- Reliability, safety, evidence, and auditability matter because the platform is connected to real operations.
| Principle | Public copy |
|---|---|
| Platform, not projects | Every deployment should strengthen reusable capabilities instead of creating a customer-specific fork. |
| Reliable data before AI | Accuracy, context, and traceability come before advanced analytics or AI recommendations. |
| Understanding before automation | The platform must understand assets, processes, constraints, and authority before it acts. |
| Human + AI collaboration | AI should assist, recommend, and execute only with the permission and scope appropriate to the role. |
| Configuration over customization | One reusable platform should adapt to factory reality without becoming a collection of separate products. |
| Every insight should create action | The operating loop should move from data to insight, recommendation, ownership, action, and outcome. |
| Physical-world responsibility | Reliability, safety, evidence, and auditability matter because the platform is connected to real operations. |
Built with factories, not at a distance.
Industrial software becomes credible when it reflects the real environment: the machines that are available, the signals that can be trusted, the operators who know the process, the constraints that shape the shift, and the managers responsible for the result. Merikh works with design partners to define a measurable starting point, validate the operational model on the floor, and turn what is learned into reusable platform capability.
What we will not compromise.
- Reliable data and transparent calculation logic
- Safety-aware, permissioned interaction with physical operations
- Clear evidence and accountability behind recommendations and actions
- Honest separation between available capability, active pilot, development, and vision
- Respect for the experience of operators, engineers, and factory teams
A cross-functional team for physical operations.
Building a Factory OS requires industrial knowledge, product thinking, data and AI engineering, connectivity, deployment, design, and close collaboration with factory teams. Team biographies and leadership profiles should appear here only after titles, experience, and photography are approved for publication.
Let’s start with the factory reality.
If you operate a factory, support industrial transformation, build production technology, or want to contribute to the future of industrial intelligence, we would be glad to hear what you are working on.