Weather Scientist (Numerical Weather Prediction)

About the role

Pravah is building the world’s first foundation model of the electric grid and seeks a Weather Data Scientist to advance next‑generation weather forecasting systems for India. You will focus on observational data quality, geospatial consistency, and integration of physics‑based models with machine learning.

Key responsibilities

• Run regional numerical weather prediction (NWP) models to generate high‑resolution forecasts and training data.
• Develop and operate a cycling data‑assimilation pipeline, including observation quality control, bias correction, and thinning workflows.
• Build and benchmark multiscale forecasting systems against reanalysis and observations, with emphasis on nowcasting and extreme events.
• Tailor weather prediction models for renewable‑energy applications such as solar GHI and 100 m wind generation.
• Collaborate with ML and software engineers to create ML‑ready global and regional weather datasets at TB scale.

Required profile

• Master’s or PhD in geophysical sciences, physics, applied mathematics, computer science, statistics, or related field (or BSc with 3+ years relevant experience).
• Hands‑on experience with limited‑area or mesoscale models (e.g., WRF, MPAS) and high‑resolution convection‑resolving simulations (~1 km).
• Demonstrated depth in data assimilation, including variational (3D/4D‑Var) or ensemble (EnKF, LETKF) methods and operational pipelines.
• Experience processing TB‑scale observational datasets (radar, satellite, radiosonde, weather stations) and using HPC environments.
• Strong communication skills to explain technical work to domain experts and cross‑disciplinary collaborators.

Required skills

• Python (xarray, dask, zarr, netCDF)
• Numerical weather prediction models: WRF, MPAS, IFS, GFS, BharatFS
• Data assimilation frameworks: JEDI/UFO, GSI, DART, PDAF
• Observation datasets: ERA5, MERRA‑2, IMDAA, IMERG/GPM, GOES/INSAT/Himawari
• High‑performance computing (HPC) and production‑grade pipeline development

What we offer

• Opportunity to develop weather forecasting models deployed for real‑time applications.
• Work on open‑ended problems at the intersection of AI and physical infrastructure.
• Close collaboration with a deeply technical, distributed team across India and the US.