Architect secure AI systems, evaluate Frontier AI models, and advance multi-agent safety frameworks across high-stakes domains.