Geospatial Data Scientist · Remote Sensing · Climate Analytics
Turning satellite imagery into climate-action decisions. 4+ years building end-to-end spatial AI pipelines for Tamil Nadu's most ambitious land-use and sustainability initiatives.
I'm a Geospatial Data Scientist based in India, specialising in turning raw earth observation data into tools that governments and planners actually use. My work sits at the intersection of remote sensing, machine learning, and climate policy.
With a background in Chemical Engineering and Petroleum Refining, I bring a systems-thinking mindset to every spatial problem — from modelling watershed hydrology to classifying land cover at regional scale using Sentinel-2 imagery and deep learning.
Over 4 years at Auroville Consulting, I've delivered end-to-end geospatial workflows for the Tamil Nadu State Planning Commission, building frameworks that directly inform land-use policy and climate adaptation strategy. My work spans renewable energy siting, blue-green infrastructure planning, and multi-criteria land suitability assessments — each producing policy-ready maps and reports that are embedded into Tamil Nadu's spatial planning framework.
Anna University, Chennai
2019–2021Arunai Engineering College
2015–2019GUVI (IIT Madras)
2021–2022LULC classification · U-Net segmentation · Random Forest · Multi-criteria suitability modelling
Policy-ready maps and reports embedded into Tamil Nadu's spatial planning framework
State-scale geospatial framework identifying, mapping, and prioritizing Blue-Green Networks (BGNs) across all 38 districts of Tamil Nadu. Processed 15,000+ Sentinel-2 images alongside 40+ geospatial layers covering LULC, hydrology, DEM, biodiversity, and climate data to deliver district-wise intervention zones evaluated on 18 ecosystem service criteria.
Deep learning pipeline using U-Net architecture on high-resolution satellite imagery to detect and map coconut plantations across Coimbatore. Achieved high precision in distinguishing coconut from mixed-use and forest land cover classes, enabling scalable crop-type mapping for agricultural planning.
Thermal remote sensing analysis of urban heat island intensity across Chennai metropolitan area. Combined MODIS LST, Sentinel-2 NDVI, and urban morphology data to identify high-risk heat zones and their correlation with LULC patterns. Deliverables included climate risk maps and evidence-based urban greening recommendations.
Multi-criteria land suitability assessment to identify optimal sites for distributed solar energy deployment across Villupuram district. Integrated slope, land cover, grid proximity, and socio-economic exclusion layers.
Modelled Time of Use electricity tariff structures for Tamil Nadu to incentivise demand-side flexibility and accelerate renewable integration. Analysis covered load profiles, peak shifting potential, and consumer impact.
Comprehensive land suitability assessment identifying unused lands for forestation, agriculture, water harvesting, housing, industrial development, and solar energy. Included LULC change detection and ground truthing campaigns.
Random Forest and SVM-based multi-class land use/land cover classification across Tamil Nadu using Sentinel-2 multi-temporal composites. Achieved >90% overall accuracy across 12 land cover classes including agriculture, water bodies, built-up, and scrubland.
Automated micro-watershed delineation from SRTM and Cartosat DEM data to identify ecological linkages and hydrological flow paths across Tamil Nadu's landscape. Foundation layer for the Blue-Green Infrastructure framework.
Multi-temporal change detection across Tamil Nadu to map fallow and unused land dynamics from 2019 to 2023. Integrated spectral indices and machine learning classifiers to assess forestation potential and identify priority intervention zones.