The Hub
Augmented product designer and practical AI tutor in Berlin.
I am an augmented product designer and practical AI tutor. I design and build digital products with AI assistance, then turn what works — and what fails — into tutorials, workshops, and training. Practice makes the teaching credible; teaching makes the practice transferable.
Practice and teaching
Code:Emotion supplies the design discipline. Human in the Middle makes the method teachable. AAA provides working proof.
Design
The design foundation: product thinking, interfaces, visual systems, and thirty years of judgment behind the products and the teaching.
AI framework
The framework and teaching practice: decide what to automate, what remains human-owned, and where review belongs. Arthur's AI Lab and HITM Labs make the method learnable.
Working proof
AAA is where the claim meets daily use: native applications conceived, designed, built, tested, and maintained with AI assistance while product judgment remains human-owned.
Credibility loop
I teach from work that has consequences beyond a demonstration. AAA provides the working record: product decisions, interface logic, model-assisted implementation, bugs, corrections, release gates, and daily use.
Arthur's AI Lab shows how I explain that practice. HITM Labs turn it into guided learning for students, teams, companies, and organizations. Together, the applications and the teaching create a productive loop: practice gives the lessons substance, and teaching brings sharper questions back into the work.
Projects
A public workbench for projects that are still moving. Each entry states what works, what does not, and what comes next.
An open-source, self-hostable design tool for campaign groups and community associations. Built with AI. Shipping without any.
A working native macOS planning app—and proof of product design authority before, during, and after AI-assisted implementation.
A working native macOS file-curation app—and a second proof of augmented product design with AI as an engineering collaborator.
Writing
Essays, notes, case studies, and practical observations on design, systems, and everyday life with AI.
Three products, three domains, one shape: the software explains, offers, records — and never decides. Including the time I shipped an approval that approved nothing.
DONE. is a concept mobile app built around one deliberately antiproductivity constraint: choose a single meaningful task, commit to it, finish it, and then stop.
A controlled study shows different commercial chatbots materially diverge in how they respond to users exhibiting delusional symptoms.
Better systems. Better judgment. More human results.
