Articles & Contributions
From agile requirements management and AI-supported requirements to agentic coding and the efficient use of AI in software development Subscribe to the newsletterArticles
AI Competence in the EU AI Act: What Article 4 Now Requires
The EU watered down Article 4 of the AI Act in July 2026. The requirement to achieve specific results was replaced by a duty to make reasonable efforts; no one is required to guarantee a certain level of competence among individual employees. What this change means for companies, what deadlines now apply, why the European Commission expressly does not require a certificate, and what this means for Swiss companies.
Agentic Harness: When AI Builds Professional-Grade Enterprise Software
An experienced software architect rebuilt an entire enterprise application using AI: microservices, tests, documentation, and true quality standards. The result: zero technical debt, 85 percent test coverage, and—depending on how you calculate it—25 to 93 times faster than a human team. The key to this is called Agent-Harness. This is exactly the discipline we’ll teach you and your team.
Setting Up Claude Code Statusline: Download the Script
A Claude code session can run smoothly for a long time, but later on, you might overlook details that were still there at the beginning. The model remained the same. Its working context has changed. That’s exactly why I built my own status bar. It shows me while I’m working...
Designing CLAUDE.md correctly: The 2026 architecture that finally makes Claude code work
Why does Claude Code feel useless to most developers? Because most only configure 1 of 4 layers. They run /init, get a generic CLAUDE.md and wonder why Claude hallucinates file structures, recommends wrong libraries and...
Agentic Coding Tools 2026: The 7 frameworks that will take your development to a new level
Forget individual AI autocomplete suggestions. The future belongs to agents that plan, orchestrate and deliver autonomously. The year is 2026, and the AI coding landscape has completely changed in the last twelve months. Away from simple code completion. Towards...
Token optimization 2026: Save up to 80% LLM costs
As you read this article, companies around the world are burning through millions in token costs - completely unnecessarily. The question is no longer IF you use cloud LLMs like Claude or GPT, but HOW EFFICIENT you do it. Because this is where the decisive competitive advantage lies...
Claude Code experience report: 16 hours instead of 6 weeks
A field report from Michael - requirements engineer, former developer, and now a bit of a tech lead again. The result first At the weekend, I built a complete application. As an MVP. With frontend, clean architecture,...
Agent harness: Why 90% AI agents fail and how you can do better
Executive Summary 90% of developers change the AI model when their agent makes mistakes. This is the wrong strategy. The AI agent is not the model. The agent is the harness - the software infrastructure that makes the model productive. This harness consists of...
The ultimate Supermaster prompt: When APEX meets Boris Cherny
The quality of your AI-supported software development stands and falls with a single factor: the system prompt. While development teams in companies work with AI IDEs such as Cursor, Windsurf or Cline on a daily basis, many overlook a fundamental fact: these...
DE 
EN