Independent AI safety researcher building SEVERANT, a formally verified, hardware-constrained AI architecture where unsafe behaviour is physically impossible by design. Working at the intersection of formal verification, hardware architecture, and AI safety. Sole researcher, no institutional affiliation.

Feedback on formal verification applied to AI constraint systems, connections to researchers working on hardware-enforced safety, and introductions to funders active in technical AI safety.

Technical discussion on formal verification in Lean 4, causal knowledge graph construction, and architectural approaches to AI safety that don't rely on training-based alignment.