Presenter

Erik Pautsch is a second-year Ph.D. student in computer science at Loyola University Chicago. His research has explored agentic approaches for HPC by building and evaluating LLM-based agent systems for education and developer support. He has also investigated software registries and evaluation protocols for agents, and is now studying how agents can guide energy-efficient code refactors and optimization choices. Previously, he worked on uncertainty-aware deep learning for scientific applications and on portability between CUDA and SYCL, including multi-summer research experience at Argonne National Laboratory.