The AI platform for Europe's industry. On top of all your systems. No data project.
The demi platform, layer by layer
One data layer over all your systems.
Your existing systems stay. demi adds data, rights and intelligence on top, and you work with all your data in one place. From the office floor to the shop floor, hosted in the EU.
How the platform worksGraphic: at the top, apps, agents, the included chat and your AI tools. In the middle, three layers: the Intelligence Layer, where AI does real work and every answer has a source; the Permission Layer, where everyone sees only what they may see; the Data Layer, with all data in one format and no duplicates. At the bottom, your systems, which stay: ERP, CRM, PLM/CAD, MES, documents, email and custom or legacy software.
Data LayerAll your data in one format, right on top of your systems.
ERP, CRM, PLM, files, email, machine data and custom or legacy software, joined into one digital twin, every value with its source. Nothing gets replaced.
How the Data Layer worksGraphic: a loose pile of ERP, CRM, PLM/CAD, files, email, machine data and custom and legacy software. The Data Layer lands on top, turns into a grid and is designed to bring your data together in one data set.
Permission LayerYour org chart becomes rights.
Every department, team and role sees only what it may. Whatever an AI wants to change, a person signs off, on the record.
How the Permission Layer worksGraphic: the org chart of a fictional company. The filters department, team and role narrow it down to one person, who sees only their part. Then an agent asks to update prices; a person signs off, and the entry lands in the record.
Intelligence LayerAgents that calculate instead of guessing.
Agents plan long tasks, check every step and pick the right model for each: fast, large, open-weight or private in the EU.
How the Intelligence Layer worksGraphic: your AI sits inside your rights. Below it three building blocks: agents for long tasks, the task runtime that plans, runs and checks every step, and demi HALO, the AI gateway that routes each step to a model class: fast, large, open-weight or private in the EU. A finished result closes it.
A click opens the film large, with captions in the picture and a transcript. Below it, the interactive tour.
- Data centres to ISO/IEC 27001
- Primary site in Germany
- GDPR, DPA in the contract
- Model training: governed by the DPA
Use cases
From company knowledge to the right pump: choose your industry and see what demi takes on for you. Use the + button to build your selection.
customer references
Small and large industrial companies already work with demi in production.
Components, machinery, plants: you connect your systems to demi.
Next webinar: Wednesday 9 December, 14:00 CET · For marketingAll dates
Your systems
No data project. Start with AI in one day.
demi brings the data from every system into one open, unified format. There is an adapter for every system, legacy systems too: read, write, sync. Your systems stay as they are.
- Start small and at low cost.
- Get going right away, with no effort for your IT.
- Your data becomes ready step by step.
- Open standard, no lock-in.
ERP
SAP S/4HANA
SAP Business One
Microsoft Dynamics 365
Infor LN
- Proalpha
- Abas ERP
- and more
CRM
Salesforce
Dynamics 365 Sales
HubSpot
SAP Sales Cloud
- CAS genesisWorld
- Pipedrive
- and more
PLM / CAD
- Teamcenter
- Windchill
- SOLIDWORKS
- Autodesk Inventor
- CATIA
- Creo
- and more
MES & machines
- MPDV HYDRA
- Siemens Opcenter
- FORCAM
- SIMATIC S7
- SINUMERIK
- OPC UA
- and more
Databases
- Microsoft SQL Server
- MySQL
- PostgreSQL
- Oracle Database
- Microsoft Access
- and more
Files & storage
SharePoint
- OneDrive
- Network drives and file shares
- DocuWare
- ELO
- d.velop
- and more
Email
Outlook
- Exchange
Gmail
- IMAP/SMTP
- and more
Legacy & in-house software
- IBM i (AS/400)
- Navision (Classic)
- Baan IV
- HCL Notes/Domino
- Excel and VBA tools
- COBOL applications
- and more
“AI knows too little to be useful,
or too much to be safe.”
Built for industry.
- annual revenue of our customers
- €2 bn+
- industries, one data layer
- 26
- until the first case runs
- ~1 day
The difference
A new approach to data, and you'll notice it every day.
Many AI projects start with a long data project. demi links your data as it is today: from every system into a digital twin of your company that learns with every use. New data you simply add later.
Nothing to structure up front
You don't have to clean up your data first. demi takes it as it is and links it by itself.
Precise answers, each with its source
With every answer you see which document, email or record it comes from.
Even large, messy data
Old file shares, duplicate entries, many formats: demi handles even very large, inconsistent data.
Permissions per person, role and document
Fine-grained and based on your org chart: the AI sees only what the person asking may see.
Data Layer
Your data universe.
Every customer, every part, every order: linked, with its source.
2,290 things. One company.
2,539 links ·
Pick a kind to highlight it.
Hover over a dot to see what it is and what it links to. Click to hold it.Tap a dot to see what it is and what it links to.
Example: fictional company · On your phone you see a section of it.
Why AI fails
When AI fails, it's not only the model. It's data, rights and workflows too.
- Data60%of AI projects without AI-ready data will be abandoned through 2026, Gartner expects.Gartner, 2025Forecast(opens in a new tab)
- Rights21%of leaders surveyed have clear rules for what their AI agents may see and do. Without such rules, an agent sees too much or may do nothing at all.Deloitte, 2026(opens in a new tab)
- Workflows84%of companies have not yet redesigned jobs and ways of working around AI.Deloitte, 2026(opens in a new tab)
The Industrial AI Operating System for Europe
Europe's industry built our prosperity, and with AI we make sure it builds the future too.
Why demi existsFAQ
Frequently asked questions
What companies most often want to know before they start.
What is demi?
demi is the AI platform for industrial companies. demi sits alongside your existing systems, brings your data together in one Data Layer, turns your org chart into permissions and only then lets AI work: traceable, and with each person's permissions. Nothing is replaced.
More in the FAQ: What is demi?How is demi different from a CRM?
A CRM stores what your team did. demi uses that data and the data from all your other systems: people and their AI agents do routine work with it, such as answering technical inquiries. Your CRM stays. demi works with each person's rights and writes back only after sign-off.
More in the FAQ: How is demi different from a CRM?Where should I start?
With a task that comes up often and takes a lot of manual work today, such as sorting inquiries or answering questions about delivery dates. Cases like these are a good start, because every company has them.
More in the FAQ: Where should I start?Do our systems stay, and does demi write to them?
Your systems stay. demi writes back only after sign-off. The Data Layer connects ERP, CRM, documents and email. Nothing is replaced, and you can go back at any time.
More in the FAQ: Do our systems stay, and does demi write to them?How quickly is demi up and running?
The first case runs in about a day, read-only at first. Nothing is replaced. You add more systems step by step, and demi writes back only after your sign-off. If you want help at the start, our services cover it.
More in the FAQ: How quickly is demi up and running?Is demi secure and GDPR-compliant?
demi runs in EU data centres. Agents and image generation work with models in EU data centres; other regions only with your company's approval. The data processing agreement under Art. 28 GDPR is part of every contract, our TOMs are public, and what applies to model training on your data is governed by the DPA (§ 3 para. 3). Where a provider has a location in the USA, EU standard contractual clauses or an adequacy decision apply. The sub-processor list names every provider and its location.
More in the FAQ: Is demi secure and GDPR-compliant?





