Brian Staber

Hi! I’m a research engineer in scientific machine learning. My background combines computational physics, applied statistics, and machine learning.

In 2018, I completed a PhD under the supervision of Johann Guilleminot in uncertainty quantification for computational mechanics, with applications to biological systems. My research focused on the stochastic modeling of patient-specific constitutive equations for soft biological tissues and surrogate models for multiscale mechanics.

Since then, I have worked as a research engineer on aeronautical and defense applications, specializing in scientific machine learning. For nearly six years, my work has centered on developing surrogate models to accelerate engineering design. I have dedicated most of my time to deep learning approaches such as graph neural networks, kernel methods, and uncertainty quantification techniques, including Bayesian inference and conformal prediction. I also have experience in industrial anomaly detection.

I am deeply curious and always eager to learn, thriving on new challenges that push me to explore different domains and methodologies. Whether it’s tackling a complex problem, experimenting with new algorithms, or diving into an unfamiliar field, I really enjoy expanding my knowledge, or at least I try to. I also enjoy developing scientific software in Python, C++, and recently Rust (we all make mistakes).