Machine Learning for Weather and Climate
I build AI and numerical-weather-modeling workflows for wind and extreme-weather prediction, with a focus on operationally useful, physically consistent forecasting tools.
Postdoctoral Associate
About
I am an atmospheric scientist working at the intersection of numerical weather prediction (NWP) and machine learning. My research spans wind gust nowcasting, precipitation forecasting, and wind-profile estimation from sparse observations and global reanalysis products.
I currently work as a Postdoctoral Associate at the University at Albany. Previously, I was a Postdoctoral Researcher at TU Delft and completed an Integrated MS & PhD at IIT Madras focused on WRF calibration and cyclone prediction.
Core tools and methods- WRF
- ERA5 / CERRA
- Mesonet and lidar profilers
- TensorFlow
- PyTorch
- Scikit-Learn
- TabNet / CNN / U-Net / SwinTransformer
- Gaussian Process Regression
- Xarray / Dask / Zarr
- HPC and multi-GPU training (NVIDIA DGX, A100 and H100)
Experience
Postdoctoral Associate, Atmospheric Sciences Research Center
-
University at Albany
Sep 2024 - Present
Researching AI-augmented weather prediction for wind and extreme events.
- Develop deep-learning frameworks for wind gust nowcasting and field reconstruction from sparse station data.
- Build scalable training/inference workflows on NVIDIA A100 multi-GPU infrastructure.
- Publish and present at venues including AMS and Gordon Research Conference.
Postdoctoral Researcher, Geosciences and Remote Sensing
-
TU Delft
Sep 2022 - Aug 2024
Led research on offshore wind resource assessment and atmospheric modeling.
- Modeled frontal low-level jets and extreme wind ramps over the North Sea.
- Estimated offshore wind power potential using gray-zone atmospheric modeling.
- Published work in Wind Energy Science, Energy, and related journals.
Institute Postdoctoral Equivalent Fellow
-
IIT Madras
Jan 2022 - Jun 2022
Advanced WRF parameter sensitivity and calibration for tropical cyclone prediction.
- Applied machine-learning-driven multi-objective optimization to improve cyclone forecasts.
- Integrated data assimilation and sensitivity analysis for robust model performance.
Education
2017 - 2022
Integrated MS & PhD, Mechanical Engineering
Indian Institute of Technology Madras
Dissertation: Sensitivity based Calibration Strategy with Data Assimilation to Improve the Prediction of Cyclones over the Indian Subcontinent.
Projects
DFS Project
ActiveAI-based statistical downscaling and census-tract-level climate physical risk assessment for New York State, funded by the NYS Department of Financial Services.
WISER Project
ActiveNSF-funded Industry-University Cooperative Research Center bridging meteorological research and the energy sector to improve power grid efficiency, reliability, and resilience under changing climate conditions.
EU-SCORES (TU Delft)
CompletedWind-resource and atmospheric modeling for integrated offshore renewable energy parks combining wind, wave, and solar, under the €45M EU-SCORES Horizon 2020 initiative.
Publications
2026
2025
Modeling frontal low-level jets and associated extreme wind power ramps over the North Sea
S2G-DI: A Deep Learning Framework for Generating High-Resolution Wind Gust Fields from Sparse Mesonet ObservationsPreprint
2024
Estimating high-resolution profiles of wind speeds from a global reanalysis dataset using TabNet
Estimating the offshore wind power potential of Portugal by utilizing gray-zone atmospheric modeling
Gaussian Process Regression‐Based Bayesian Optimisation (G‐BO) of Model Parameters—A WRF Model Case Study of Southeast Australia Heat Extremes
2023
2022
Determining the sensitive parameters of the Weather Research and Forecasting (WRF) model for the simulation of tropical cyclones in the Bay of Bengal using global sensitivity analysis and machine learning
Impact of data assimilation on a calibrated WRF model for the prediction of tropical cyclones over the Bay of Bengal
2021
Achievements
NWO EINF Grant (2024)
Awarded 1 million SBUs for EU-SCORES solar-wind hindcast research.
Talks & Conferences
Contact
I am open to collaborations in AI for weather and climate, wind-energy analytics, and applied atmospheric modeling.