Topics I'm currently interested in: - Distributed training of large neural networks - Scaling laws for robotics - Marginal risk evaluations for AI systems - Mechanistic interpretability at scale If you are working on these, please feel free to drop me a note. My email is easily discoverable online