KNOWLEDGE MANAGEMENT
Knowledge management with AI: your company knowledge, reviewed and with sources
Bring handbooks, notes and experience together in knowledge spaces. Your agents find relevant content and cite the right source – and whatever the AI learns only counts once your team has approved it.
Knowledge spaces instead of scattered files
- Notes in folders, as Markdown – they remain your files.
- Notes link to each other with double square brackets; “Connections” shows the whole network and links that lead nowhere.
- Obsidian-compatible import and export.
- Every earlier version is kept in the history.
- Agents read your notes, work with them and quote from them.
The AI learns – your team decides what counts
- New insights, open questions and contradictions land under “Review” – you see exactly what changes.
- The person responsible approves, edits first or rejects.
- The Learning centre shows how your knowledge is used; the maintenance run finds gaps, and test questions check whether the right note is found.
- AI tools summarise, create an FAQ or a glossary and check for contradictions – every result lands under “Review” first.
Examples from practice
- Manufacturing: the shift assistant makes approved fault information easy to find.
- Onboarding: the onboarding buddy answers recurring questions from approved knowledge; if something is missing, the gap becomes visible.
- IT service: proven solutions from resolved tickets become runbook knowledge.
Transcript
Your knowledge spaces describe how you work. On the left are your notes, sorted into folders – like the basics, from ideal clients to services and pricing. Everything is Markdown and stays your own file. Notes link to each other with double square brackets. Agents read them, work with them and quote from them. Use the pencil to edit a note – and every earlier version is kept in the history. “Connections” shows the whole network: which notes belong together – and where links lead nowhere. You add something new as a note – or you save an insight. It only becomes part of your knowledge once it’s been reviewed.
Transcript
Knowledge only stays good if you look after it. Under “Review” are the proposals from agents and workflows – for example a new weekly plan or an addition to a note. You see exactly what changes. Then you approve it, edit it first – or reject it. The Learning centre shows how your knowledge is used. The maintenance run finds gaps, and test questions check whether the right note is found. The AI tools summarise, create an FAQ or a glossary and check for contradictions. Every result lands under “Review” first.
How your knowledge stays well maintained
- Create or import notes – from Obsidian, for example.
- Agents answer with sources; you save good answers as insights.
- Review and approve proposals – only then do they feed into answers.
- Keep an eye on quality with test questions and the maintenance run.
Frequently asked questions
How does Obsidian fit in?+
The knowledge is stored as Markdown. Notes can be imported and exported, and the storage is Obsidian-compatible. Approvals, version history and knowledge checks complement this open knowledge base.
Can the AI change the knowledge by itself?+
No. Changes are proposals; your team approves, edits or rejects them.
Who can see what?+
You assign roles in the team – admin, editor, member or reader – with access per knowledge space.
More topics
See DENKRAUM AI with your own example
In a 45-minute demo, we show you suitable examples. Then we define a first use case together, along with the knowledge sources needed and the operating concept.