Philosophy, linguistics, and German Studies at HHU Düsseldorf. The practical insight I took from it: meaning isn't in words, it's in use — and that turns out to be the foundational problem in agentic engineering, where agents fail on thin context, not thin capability.

Most of what matters in a deployed AI system isn't the model. It's the harness: the scaffolding where expertise gets encoded as infrastructure. Tool surfaces, memory architecture, orchestration patterns, the governance layer that determines what runs autonomously and what requires a human — these are the surfaces where domain knowledge, operational judgment, and accumulated experience become something agents can act on. I study and build at this layer from first principles, because the encoding is where the real work happens and where the real differentiation lives.

Organizations running on agentic infrastructure don't just automate — they restructure. The agentic OS is real: machine-readable operating manuals, expertise encoded as infrastructure, agents executing while humans govern. The coordination layer doesn't disappear. It relocates into what I call the instantiation: the specific, irreducible way an organization deploys intelligence against its actual problems.

The blog covers what I find at this layer — agent harness design, context engineering, the SaaSpocalypse, agentic commerce, and the emerging shape of agentic organizations. Practitioner-positioned, research-grounded.