At NVIDIA, we are building DGX Cloud, which offers cloud-based AI-as-a-service on optimized GPU accelerated hardware and software stacks. My team is responsible for the core compute software that schedules and runs scalable AI, Deep Learning and HPC multi-node workloads on GPU container runtime systems. Read more about it here: https://www.nvidia.com/en-us/data-center/dgx-cloud/Previously, I've worked as a research scientist and contributed to multiple DoE/DARPA funded research and advanced technology projects for exascale systems, specifically in areas of novel programming models, parallel languages, compilers, and runtime systems.My PhD dissertation titled "Runtime Systems for Extreme Scale Platforms" identified the critical role of programming models and scalable runtime systems to handle the challenges of exascale computing.Read more about it here: https://hdl.handle.net/1911/76173Over the course of my career, I have developed skills to design scalable high-performance computing systems involving diverse processor architectures. I have co-authored more than 15 journal, peer-reviewed conference and workshop papers, and presented more than 10 invited talks, posters and demos. Contributed to open-source projects and written programming model specification documents. Served as general chair, session chair and program committee member at multiple technical conferences and workshops, as well as mentored students at both senior and graduate level research projects.
Listed skills include Algorithms, C++, Embedded Systems, Linux, and 4 others.