KI Grundlagen lernen für Einsteiger
Was ist ein KI Modell und wie lernen Maschinen? Der Guide.
What an AI model actually is and how machines learn — the jargon-free starting point.
Resources
Audiobooks, blog and videos on AI
Everything I have published on working with AI, free to read or a few francs to listen. Start with the fundamentals, go deep with the series.
Getting started
Three short audiobooks for anyone who wants to understand AI and use it well. No technical background needed.
KI-Prompting: Praxiswissen für Anfänger
KI Ergebnisse präzise steuern und KI produktiv nutzen
Practical prompting for beginners — the fundamentals for getting genuinely useful results from AI.
KI-Agenten: Praxis, Skills und Kontext
Lerne Prompting und KI Kontext für den Alltag.
A hands-on guide to building with coding agents: skills, context, and real workflows.
Series: Vom Prompt zum Loop
Eight parts, 35 to 39 minutes each, on the evolution of AI control: from a single prompt to autonomous, verified agent systems. Listen in order.
-
Part 1: Das Handwerk der Worte
Prompt Engineering, System-Instruktionen und die Kunst klarer Anweisungen
Roles, few-shot examples, chain of thought and system instructions: prompting as a craft, not a guessing game.
-
Part 2: Kontext ist König
Context Engineering, Memory und Retrieval-Augmented Generation verstehen
How to give a language model the right knowledge: context windows, memory and retrieval-augmented generation.
-
Part 3: Die Entstehung der Agenten
Tool Use, Function Calling und das Model Context Protocol
From answering machine to actor: tool use, function calling, ReAct and the Model Context Protocol.
-
Part 4: Das Harness und die Umgebung
Harness Engineering, Leitplanken und automatische Evaluatoren
The environment decides whether an agent is production-ready: guardrails, sandboxes and automatic evaluators.
-
Part 5: Die Geburt von Loop Engineering
Vom einzelnen Prompt zu autonomen Arbeitszyklen
Plan, execute, observe, refine: why autonomous work cycles beat single prompts on every long task.
-
Part 6: Anatomie eines perfekten Loops
Abbruchbedingungen, Kontextmanagement und Memory in autonomen Schleifen
Stop conditions, context rotation, memory and a human emergency stop: what keeps an autonomous loop safe.
-
Part 7: Verifizierung und Self-Correction
Testgetriebene KI-Entwicklung, Kritik-Schleifen und Fallbacks
A model must not grade its own work: generator-evaluator loops, hard tests and fallbacks for AI you can trust.
-
Part 8: Das autonome Ökosystem
Multi-Agenten-Orchestrierung und die Zukunft der KI-Arbeit
Teams of specialised agents, hierarchy versus swarm, and the human as architect, auditor and conductor.
For developers
The longer career guide for people who build software.
Vom Entwickler zum KI-System-Experten
Programmieren mit KI als Karriere-Booster nutzen
How to stay relevant when agents write the code — the skills that survive the AI wave.
All audiobooks are in German, read by a digital voice, available on Spotify, Apple Books, Google Play and other stores.
On YouTube
The first parts of the series and free first chapters of the audiobooks, as videos.
From the blog
Notes on AI engineering, coding agents and software that holds.
-
AI Agents in Daily Development: What Actually Works
Coding agents deliver impressive demos — and often fail in real project work. What survives sustained practice: three patterns that hold up, and three mistakes everyone makes.
-
From Writing Code to Designing Systems: How the Developer Role Is Shifting
If agents write the code, what is left of the developer profession? More than the headlines claim — but in a different place. An honest assessment.
-
AI Prompting: The Five Mistakes Almost Every Beginner Makes
Mediocre AI results are rarely the model's fault — usually it's the prompt. Five typical beginner mistakes and what works instead, with everyday examples.
From listening to building
The audiobooks explain the ideas. In the Kickstart workshop your team builds its first tool with AI in one day, for a real workflow from your business.