Claude training for companies in Munich

Why industrial companies around Munich train their teams on Claude

KfW Research reported in February 2026 that 20 percent of Mittelstand firms used AI in 2022 to 2024, and 36 percent of those with 50 or more employees (KfW Research). KfW ties AI use to expertise in house. In an industrial company that expertise sits with the people who write specifications and answer suppliers, and they rarely have time to teach themselves a new tool.

A works council asks two questions about any new tool: what it may do, and who checks its work. The training answers both before the first session, and the list further down collects the questions a works council is likely to raise.

What a Claude training for a Munich team covers

The content follows the documents your team handles. The parts for an industrial team:

• Projects per product line: a project holds the specification template, the terms your company uses and the standards it cites, so every chat starts with the right context.
• Instructions with a fixed check: the team writes down how a supplier letter is drafted, what Claude may fill in and what a person compares against the drawing.
• Skills for repeated documents: a written procedure for a test report or a deviation notice that Claude loads when the task matches.
• Connectors: Claude reads from your document management or your ERP through MCP servers, so no one pastes part numbers by hand.
• For developers and automation staff, in their own session: Claude as coding agent under rules that a hook enforces before each tool call.

Exercises for industrial teams: a requirements list from a customer’s specification, a supplier letter in German and English from one deviation report, a summary of a long standard for the sales team, and a checklist that compares a quotation against the request.

Online from Frankfurt, or on site at your plant

I work from Frankfurt. Online sessions run by video and suit companies with several sites, for example a head office in Munich and a plant in Augsburg or Ingolstadt. For an on-site day I come to your site. Participants work on their own laptops with your company’s Claude accounts, and the shop floor is close enough to fetch a real document when a question comes up.

Both formats are built on your company’s documents. The needs assessment decides which format fits.

What the works council will ask

Questions a works council asks about an AI tool, which the needs assessment answers before the first session:

1. Which tasks does Claude take over, and which stay with a person?
2. Which data goes into Claude, and where is it stored? Your data protection officer and the settings of your Claude plan answer this.
3. Does the tool log who entered what, and who can read that log?
4. Is participation voluntary, and what happens for staff who do not take part?
5. Does the training measure individual performance? It does not; the list of participants names topics, not results.
6. Who answers questions after the training, and how are they collected?
7. Are the documents used in the exercises anonymized, and who approved them?
8. When is the content reviewed again, for example after a change of the Claude plan?

Who I am

I am Michael Wutzke, an IT architect for data platforms in Frankfurt with more than twenty years in IT and media. I hold a technical diploma in mechanical engineering and English. I chose it at 20 because I wanted to build machines; today I build software. I teach at Claude Hacker House, hold a training authorization from IHK Frankfurt for two IT apprenticeships, and I take part in Claude Certified Architect Foundations, Anthropic’s first certification program. I am also interested in open-source AI models that a company runs on its own servers. Details: Teaching and certifications.