Raphaël Pestourie

Raphaël Pestourie is an Assistant Professor in the School of Computational Science and Engineering at Georgia Tech (2023–present). He holds a PhD in Applied Mathematics (Harvard University, 2020) and an MBA from ESSEC. His research focuses on scientific machine learning and inverse design in electromagnetism, combining AI with physics-based models for engineering optimization. Notable contributions include fast approximate PDE solvers, surrogate models for metasurface design, and end-to-end optimization frameworks.

**Education**:

• PhD in Applied Mathematics, Harvard University (2020)
• Secondary field in Computational Science & Engineering, Harvard
• MBA, ESSEC
• Masters in Nanosciences (Paris Saclay), Engineering (École Centrale Paris), and Statistics (Harvard)

**Research Interests**:
His work bridges applied mathematics and AI to solve large-scale engineering problems. Key areas include:

• Inverse design of metasurfaces for optical applications
• Scientific machine learning for PDE-based optimization
• Hybrid models combining theory-driven and data-driven approaches
• End-to-end optimization of electromagnetic systems

**Awards & Support**:

• Funded by MIT-IBM Watson AI Lab, DARPA, Simons Foundation, and others
• Arthur Sachs Fellow (French Fulbright Commission)
• Member of Harvard's Graduate School Leadership Institute

**Advisees & Collaborations**:

• Advises PhD/MSc students at Georgia Tech and collaborates with MIT, Yale, and UC Berkeley researchers
• Current projects include metasurface design for chemical detection and imaging

**Teaching**:

• Created new courses: CSE 8803 (Scientific Machine Learning) and CSE 8801 (Linear Algebra/Probability/Stats)