Hari Sadasivan

Hari Sadasivan is Member of Technical Staff at AMD where he explores performance optimizations for GPU GEMMs as part of AMD’s open-sourced Composable Kernel library team. Hari also leads/co-leads several research collaborations with Universities, in AI and HPC (-omics), where he mentors students to port and optimize software on AMD GPUs, thereby delivering open-sourced GPU solutions to advance the field. Hari designed and teaches a ROCm/HIP GPU Programming course at UW Seattle. Hari holds a PhD in CSE from the University of Michigan Ann Arbor where he worked on HW-SW solutions for accelerated and portable DNA sequencing. Prior to joining AMD, Hari has worked at NVIDIA and Samsung Research.