Welcome to Quality Booster

I’m Andi, a Quality Engineer at EnBW. For over 20 years, I’ve been asking one question: does this software actually do what it’s supposed to? For a while now, I’ve been using AI agents in my daily testing work — and it’s changing more than I expected.

I started this blog because I couldn’t find what I was looking for: honest, practical accounts of what AI really brings to testing. Not the polished demos, but the everyday reality. What works, what fails, and why.

What you’ll find here

• Practical reports — what worked in my day-to-day project work and what didn’t. No sugarcoating.
• Guides — concrete step-by-step tutorials for tools like Claude Code, Playwright, and MCP integrations.
• Tool assessments — my honest opinion, no affiliate marketing.
• Mistakes and dead ends — you usually learn more from those than from success stories.

Why now?

AI assistants like Claude and GitHub Copilot are fundamentally changing how we develop and test software. But most reports stay on the surface. I’m interested in the uncomfortable questions:

• How do I test code written by an AI — code I might not fully understand?
• Can an AI agent actually write meaningful test cases, or does it just produce tests that turn green?
• Where do I genuinely save time — and where am I just creating new complexity?

I have answers to some of these questions by now. Others, not yet. I share both here.

What I promise

I write about what I’ve actually experienced. If something didn’t work, I’ll say so. If I’m unsure about something, I’ll say that too. I’m the type who checks three times before calling something “done” — and that’s how I write as well.

If you’re in testing yourself and wondering what AI agents can really do in everyday work: have a look around. I’d love to hear from you.