Presenter

Manikya Swathi Vallabhajosyula is a sixth-year Ph.D. candidate in Computer Science and Engineering at The Ohio State University, advised by Dr. Rajiv Ramnath. She is a researcher at the NSF AI Institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment (ICICLE), where her work focuses on AI-driven optimization of high-performance computing (HPC) systems within the Middleware and Tools thrust. Her research addresses the challenge of poor resource utilization in HPC environments caused by inefficient configurations and limited system awareness. She developed HARP (HPC Application Resource Predictor), a machine learning-based framework that predicts walltime and resource requirements using historical workload patterns. She also contributed to the design of a smart scheduler for deep learning workloads, enabling dynamic, context-aware orchestration of scientific jobs.
Her work bridges traditional system-level scheduling with domain-specific computing needs in AI workloads. She aims to build adaptive, explainable, and scalable infrastructure to support AI-driven discovery across scientific fields.
Beyond research, She has mentored undergraduate and graduate students, led a 200-student summer research program in India, and organized K–12 AI outreach. As ICICLE’s NextGen Student Group Manager, she coordinates student-led initiatives and cross-institutional collaborations to foster inclusion in research.