Speakers: Jon Coons, NVIDIA and Julian Uran, PSC
Abstract: NVIDIA AI Enterprise (NVAIE) is a suite of software tools, libraries, and frameworks that accelerate and simplify the development, deployment, and scaling of AI applications. As part of their contribution to the National AI Research Resource (NAIRR) pilot project, NVIDIA is making NVAIE available to researchers and educators who utilize Bridges-2 H100 resources. In this webinar, we will introduce the NVAIE framework, including BioNeMo for drug discovery, PhysicsNeMo (physics-bound AI framework), and Cosmos for robotics and autonomous vehicles. NVAIE deployment on Bridges-2 H100 will be demonstrated by running the AlphaFold2-Multimer inference microservice, as an example of the applications that should be of interest to our community.
Bios: Jon Coons is a key member of the NVIDIA AI Enterprise software team, serving as a Senior AI Product Specialist for Generative AI and AI Inferencing for the Americas. Prior to NVIDIA, Jon was an integral part of the AI Global Black Belt team for Microsoft Azure, working with primarily Fortune 500 Enterprises on Generative AI, Vision/Video Analytics and Edge AI initiatives. Bridging the divide of technical acumen and foundational business logic, Jon leverages his diverse background to bring a unique and valuable perspective to the Generative AI equation.
Julian Uran is a Senior Systems Software Engineer at Carnegie Mellon University’s Pittsburgh Supercomputing Center, where he works on the AI and Big Data team. With a master’s degree in systems and computer engineering, Julian is a go-to expert for all things tech—software development, advanced troubleshooting, DevOps, virtualization, system tuning, and hardening. He plays a key role in supporting cutting-edge research by helping users maximize the performance of HPC systems like Bridges-2. At this event, Julian will demonstrate how to run workflows with NVIDIA NIM containers on Bridges-2’s H100 GPUs, making AI at scale more accessible to the research community.