Connect sources
DMS, drives, wiki, specialist systems.
RAG / knowledge systems: Make company knowledge usable – with sources.
Make company knowledge usable for AI with RAG: questions about your own documents are answered with sources, respecting permissions and keeping content current.
In short: RAG (retrieval-augmented generation) finds the relevant passages in your documents for a question and lets the model answer from them – with sources. The model isn’t trained; the knowledge stays in your documents.
What the building block does
Tasks and limits.
Prepare
Split, structure, assign permissions.
Search
Semantic and exact – with permission checks.
Answer
From the passages found, with citations.
Measure
Retrieval quality, gaps, how current the content is.
Maintain
Detect and replace outdated documents.
When does it fit?
Decision guide.
| Fits when | Doesn’t fit when |
|---|---|
| knowledge exists in documents but is hard to find | the knowledge isn’t documented anywhere |
| permissions are clear | nobody can say who may see what |
| documents are maintained | the sources stay out of date |
In action
What working with it looks like.
Sample images from our visual library – the interfaces shown are illustrations, not client data.
Use cases
Where this building block is used.
Technology & products
What’s behind it.
Products only where they fit the task.
RAG Engineering
How we build knowledge systems: sources, index, retrieval, permissions, measurement.
View detailsKnowledge Assistant
Search internal documents by asking a question; every answer with a source.
View details
Frequently asked questions
Frequently asked questions: RAG / knowledge systems
How is this different from a chat with file upload?
Permissions, current content, measurement, integration and operation. An upload chat is a test; a knowledge system is an operational tool.
Does RAG run locally?
Yes – that is the standard case for company knowledge (Private AI).
How do we keep it up to date?
Sources are synchronised continuously; Studio shows gaps and outdated documents.
Next step
Check RAG / knowledge systems for your process.
The AI potential assessment checks this on a specific workflow – free of charge, with a clear recommendation.