Griffin Lacey will discuss how hardware accelerators like GPUs support modern AI/ML workloads, comparing different architectures and exploring the role of the software stack and AI platforms. He will also explain how AI/ML infrastructure scales from embedded systems to supercomputers for training LLMs with trillions of parameters.
Griffin Lacey received his BEng in Engineering Systems & Computing from the School of Engineering at University of Guelph in 2014. Griffin also earned his MASc in 2016 at UoG and is currently Manager, Solutions Architect at NVIDIA.