What I'm up to:
What I was previously up to: AI infrastructure for self-driving cars and humanoid robots at Tesla. I worked on our distributed training stack; software for & bringup of custom training silicon; performance & reliability of workloads (training neural nets, offline inference, dataset generation pipelines, etc.) in our supercomputing clusters; cluster health & research enviornment management; and more related things.
What I'm interested in: Anything related to machine learning, systems software, and compute.