Gregorio von Hildebrand

Meet Gregorio

2 following1 follower

I build AI agent systems that run unattended — and that a human can still read, audit and correct mid-run. My method is Interpretable Context Methodology: the folder structure is the architecture, so nothing hides in framework code. Recent: an EU AI Act Article 50 auditor, AI oversight infrastructure, an education platform in production. Next.js, Supabase, Python, Claude Agent SDK.