My current research centers on inverse problems governed by partial differential equations, with a particular focus on thermo-fluid systems. I use machine learning, deep learning, and other data-driven techniques to address these challenges. So far, my work has focused on physics-informed machine-learning-based approaches for accurate parameter estimation in complex physical systems under sparse and noisy data. These studies provide the foundation for my ongoing and future work on real-time thermal monitoring and the development of digital twins for complex thermo-fluid systems. I have also applied similar ideas to subsurface characterization in geophysical contexts. In some of these applications, I adopt a Bayesian framework for parameter estimation, which allows for explicit quantification of uncertainty arising from noise and limited data.

Although my current research focuses on inverse problems and data-driven approaches, my primary training is in numerical modeling, the development of high-order numerical schemes, large-scale simulation, and high-performance scientific computing. I have worked on several complex systems, including flow MRI, computational electrodynamics, and magnetohydrodynamics. This background has shaped my interest in developing highly accurate and stable numerical methods for obtaining high-fidelity solutions to a broad range of complex science and engineering problems.