At present, I am leading the development of Meta’s software stack and best practices to support deployment of accelerators at scale to support over a billion users. I’ve led and/or personally developed the APIs for accelerator deployment at scale, for model performance optimization, automated inference hyperparameter optimization, model performance and correctness testing, and Meta’s environment for continuous integration and deployment of AI models on GPUs and accelerators. I’ve taken meta from 0 => many of the largest model’s in our industry deployed on GPUs and accelerators.

Specialties: transformational technology leadership, people leadership, deep learning, artificial intelligence, project leadership, strategy development, development execution, CPU and system architecture, server architecture, virtualization architecture, multiprocessor systems, supercomputer systems, compilation technology and hardware/software co-optimization.