Anirudh K M

About Me

Growing up, the monsoon wasn’t just a weather pattern; it was something you felt. The air turned heavy, the earth cracked with thirst, and then, suddenly, everything came alive with rain. Now, as a Ph.D. researcher at IIT Delhi’s Centre for Atmospheric Sciences, I chase those same feelings, but with equations and data instead of childhood awe.

My work sits at the crossroads of storms and algorithms, where I’m trying to combine the strengths of atmospheric physics and machine learning. I aim to build physics-guided neural networks to forecast the active and break phases of the Indian Summer Monsoon. To capture its complex, multi-scale behavior, I draw on deep learning architectures such as transformers and Fourier Neural Operators, aiming for forecasts that are not only accurate but also physically grounded and computationally efficient.

A central goal of my research is to better understand the interactions between monsoon intraseasonal oscillations and large-scale climate drivers like ENSO. By embedding physical constraints into models, I aim to improve their accuracy, interpretability, and efficiency, delivering forecasts that support agriculture and water resource planning.

Beyond research, I’m drawn to literature, politics, and cinema, and enjoy exploring the intersections of science, society, and culture.