Why you should really use Copilot and Langdock
Use the four-quadrant matrix to assess AI applications: improve existing work, build skills and invest usable capacity in new services. Includes examples and five decision questions.
Knowledge Hub
Understand AI. Apply it sensibly in your company.
Choose the entry point that fits your question: concrete work in the business or the technology behind the systems. Both paths lead to verifiable next steps.
YOUTUBE
Two paths through the videos
All videos are in German. Follow one path or enter at the question you need to solve now.
Apply AI in the company
Move from a concrete work situation to a verifiable process of your own.
Watch this path as a playlist (opens in a new window)- 01Read the article and matrix (German PDF)
Everyday vs. Game-Changing AI: Der 4-Quadranten-Plan
Maps AI applications by value creation, coordination and their effect on the business model.
Published Sep 21, 2026Watch directly on YouTube (opens in a new window) - 02Check your own process with the Bottleneck Check
Claude Desktop im Einsatz: Angebote automatisiert erstellen
Shows one bounded workflow instead of a general tool demo.
Published Aug 11, 2026Watch directly on YouTube (opens in a new window) - 03Discuss a concrete process in the Outside View
Automatisiere Logik, entfalte Empathie
Separates automatable logic from human judgement and relationships.
Published Aug 20, 2026Watch directly on YouTube (opens in a new window)
Understand AI technology and behaviour
Understand what the system can do and where it becomes an operational decision.
Watch this path as a playlist (opens in a new window)- 01Read the companion article “Everything is code”
Claude Code, Hermes & Co.: Die Revolution für deine Arbeit ohne selbst zu coden
Explains why coding agents can also work with documents and office work.
Published Sep 18, 2026Watch directly on YouTube (opens in a new window) - 02Understand the intelligence layer
Warum ChatGPT im Unternehmen allein nicht reicht
Shows why a chat window is not yet a reliable operating architecture.
Published Sep 2, 2026Watch directly on YouTube (opens in a new window) - 03Watch the offer demo as an application (opens in a new window)
Der Denkfehler, der deine KI-Antworten unvollständig macht
Shows how the wrong decomposition leaves answers incomplete.
Published Aug 4, 2026Watch directly on YouTube (opens in a new window)
Articles and worksheets
Filter by
The 8 Fundamental Weaknesses of LLMs
Why ChatGPT, Claude and Co. are not production-ready on their own - and what you can do about it.
Knowledge Graphs vs. Vector RAG: The Fundamental Difference
When graph methods beat classic RAG, when they do not, and what the benchmarks actually measure.
Everything Is Code: How Claude Code Makes Your Work Checkable
Claude Code and Codex without programming: why almost every file is text underneath, how a folder makes work checkable, and a prompt along the IMPACTS protocol.
The Aprixity Thesis: 2026-2030
Our scientific forecast for the future of AI-First transformation. Why determinism and transaction velocity determine success.
MECE for LLMs: Structuring Principle for Precise AI Prompts
Learn how the MECE principle from consulting structures your LLM prompts and improves AI outputs.
The 5 Dimensions of a Complete AI System
A MECE framework for evaluating and planning AI architectures. What capabilities does a system need to be truly intelligent?
From Gut Feeling to Graph: How to Automate High-Touch Processes Without Losing the Human Element
A case study from Premium Real Estate. Why traditional automation fails with complex, relationship-intensive processes – and how a semantic Intelligence Layer solves this.
What is GraphRAG? The Hybrid Approach Explained
How GraphRAG connects knowledge graphs with RAG, where the advantage is measurable and where it is not.
MCP Explained: The Model Context Protocol for LLM Integration
MCP is the open standard for LLM-to-tool communication. How it works and why it matters.
AI-First does not mean AI-Only.
AI-First is not a technology add-on. It is an order: understand, remove, simplify, accelerate, automate.
The Intelligence Layer connects data to decisions.
A strong model without dependable context remains blind. The Intelligence Layer makes knowledge structured, navigable and verifiable.