Wind and Climate Data Analyst
Wind Pioneers · Dec 2022 – Present
Build and run the reanalysis data layer: ERA5 and MERRA-2 pipelines, a shared data library, and wind-resource and hazard analysis for project sites.
Senior Associate → Climate Modeling Analyst
Genesis Ray Energy · Feb 2021 – Dec 2022
Built solar hindcast and machine-learning bias-correction models, then supervised a small team running bias correction in production on Azure.
M.Sc. in Atmosphere and Ocean Sciences
IIT Bhubaneswar · 2018 – 2020
Thesis on how tropical waves modulate extreme monsoon rainfall over India, and how that changed over four decades.
Data Collection and Assimilation (internship)
NARL, Gadanki · 2019
Field observations and an introduction to data assimilation.
Ocean Data Retrieval Techniques (internship)
INCOIS, Hyderabad · 2019
Retrieving and processing ocean observation data.
Extreme Weather and Climate Analysis (internship)
IMD, Bhubaneswar · 2018
Analysing extreme weather and climate events.
A Python library, command-line tool and desktop app that serves ERA5 and MERRA-2 reanalysis from three storage backends through one interface.
Python · xarray · Zarr · NetCDF · BigQuery · Google Cloud Storage
code private — employer-owned
Extracts 50 m New European Wind Atlas maps for a project area in about a minute, and showed why the coarse map misleads.
Python · New European Wind Atlas · GDAL · QGIS
code private — employer-owned
A tiny macOS background app that flies an airplane across your screen five minutes before every Outlook meeting.
Swift · EventKit · AppKit · launchd
· 3 min read · Wind Resource · GIS · Resolution
· 4 min read · ERA5 · Data Quality · Pipelines
· 3 min read · Migration · Testing · Extreme Wind
I'm an atmospheric scientist (M.Sc. Atmosphere & Ocean Sciences, IIT Bhubaneswar) with 5+ years turning reanalysis and forecast data into products teams can trust: production ERA5 and MERRA-2 pipelines, wind-resource and climate-hazard analysis, and machine-learning bias correction.
Most of what I care about is reliability. Upstream feeds don't always behave, so the pipelines I build check their own output and say so when something is wrong, instead of quietly passing gaps downstream.
Outside work, I'm usually behind a camera or with my PS5, or cooking or playing with my pet cat Spot.