"You can't connect the dots looking forward; you can only connect them looking backwards." Visiting Stanford has been a goal since day one of my PhD, and it is exciting to see those dots align. Honored to receive the France–Stanford Center Visiting Student Researcher Fellowship to join Prof. Beverley J. McKeon's group (Oct.–Nov. 2026). We will focus on explainable AI (SHAP) to quantify coherent structures in unsteady turbulent boundary layers. Stay hungry, stay foolish.

Presented "Navigating the Intermittency: A Generative Surrogate for Long-Horizon Forecasting of Minimal Flow Unit from Sparse Measurements" at EUROMECH Colloquium 665 on Data-driven active control in flows.

Our latest work published in Physical Review Fluids extends the framework initially presented at AiFluids 2025 with deeper analysis of the learned latent space dynamics.

The VIVALDy preprint is now available on arXiv. The framework combines a β-VAE-GAN with masked convolutions and a bidirectional transformer to reconstruct turbulent flow around a moving cylinder using only cylinder displacement as input.

An extension of the work presented at EUROMECH 629 was presented at the 11th International Symposium on Turbulence Heat and Mass Transfer (THMT-25) in Tokyo, Japan. A deeper analysis of latent space forecasting was carried out, showing the framework accurately predicts latent space dynamics and preserves chaotic characteristics within one Lyapunov time.