I'm a research coordinator and part of the team running AI Safety Camp. I'm interested in field building in the AI Safety space, including getting more competent organizers into the field and supporting outreach work to inform the public about AI risks.
My research interests focus on both Simulator Theory of LLMs as per janus, and on how to design more inherently interpretable yet flexible neurosymboloc cognitive architectures, backed by strong Agent Foundations.
My background is in Cognitive Science, including computational mathematics, neural networks, and philosophy of mind, as well as a thorough engagement with the MIRI perspective on aligning superintelligent systems.