I lead research projects focused on evaluating safety and stability in complex AI systems. My work emphasizes system-level risk, feedback dynamics, and multi-agent interactions, with the goal of producing practical, public-interest evaluation frameworks.