Joshua R. Tempelman, Ph.D.

I am a Director’s Postdoctoral Fellow in the Space Remote Sensing and Data Science Group (ISR-6) at Los Alamos National Laboratory. My research sits at the intersection of nonlinear dynamics, wave propagation, and scientific machine learning, with the goal of building trustworthy digital twins that connect physical models, real-world data, and autonomous decision-making. Before coming to Los Alamos, I earned my Ph.D. in Mechanical Engineering from the University of Illinois Urbana-Champaign in 2024, where my dissertation on passive energy management in acoustic and dynamical systems received the Stanley I. Weiss Outstanding Thesis Award. My doctoral work advanced nonlinear wave–structure interaction, topological mechanics, and targeted energy transfer, combining theory, computation, and experiment. Alongside this work, I collaborated with Los Alamos on sensing and machine-learning approaches for additive manufacturing and ultrasonic nondestructive evaluation.

At Los Alamos, my research focuses on interpretable AI for data-limited inverse problems. I develop simulation-informed generative models and reduced-order dynamical frameworks for applications including ultrasonic weld inspection and hyperspectral imaging. A central theme of my work is integrating governing equations with machine learning rather than replacing physics with black-box models. This approach enables uncertainty-aware inference, out-of-distribution robustness, and computational efficiency suitable for real-time deployment.

Looking ahead, I am building a research program centered on reliability-aware digital twins for addressing problems in advanced manufacturing and enabling AI-accelerated metamaterial design. My long-term vision is to create closed-loop, autonomous workflows that unify simulation, sensing, and optimization to support resilient infrastructure, advanced manufacturing, and intelligent mechanical systems.