About

Research scientist specializing in AI for science in NASA’s Science, Implementation, and AI teams. Alongside, I have contributed to AI strategy at NASA and IEEE and have a growing focus on frontier AI governance.

AI for Science
My projects are primarily centered around unsupervised and self-supervised learning, change and anomaly detection, time-series analysis, AI foundation models, and generating analysis ready datasets using AI/ML. These translate to a variety of research applications related to analysis of high-dimensional Earth and Space Science datasets for detecting, understanding and forecasting changes, natural hazards, energy infrastructure monitoring, and extracting rare class signals for scientific discovery.

As a NASA Black Marble Science Team member and PI a large part of my work has focused on tailoring AI/ML algorithms for studying the Earth at Night from a large, unlabeled daily satellite data stream. The derived inferences have a broad variety of applications ranging from monitoring nighttime fires, gas flares, volcanoes, to tracking urbanization, electric grid access, and reliability, to power outage mapping for disaster impact and recovery analysis. As a NASA FireSense Implementation team PI, I engaged with inter-agency stakeholders and contributed to AI-derived insights for wildfire management. I served as a Co-I in NASA’s multi-institution collaborations for developing and finetuning AI foundation models from satellite data for land and weather applications; and red-teamed the downstream use cases for operationalization.

My research has produced AI-derived data products and models — used in NSF NAIRR pilot data resources and by respective NASA Science teams for operational use, and in growing new external partnerships.

AI Strategy and Governance:
I served as an AI subject matter expert with the NASA SMD AI/ML working group from 2020-2024, co-conducted agency-wide surveys, co-hosted workshops as a focus area lead, convened yearly meetings centered around cross-divisional applications of AI, and contributed to working group reports. I also led the working group’s AI training dataset generation and benchmarking effort for the Earth Science use case. From 2022-2023, I contributed to AGU-NASA workshops for ethical use of AI/ML in Earth, Space, and Environmental Sciences. I served as a technical co-lead and social media lead in the IEEE Geoscience and Remote Sensing Image Analysis and Data Fusion Working Group from 2021-2023, co-organizing the first IEEE international computer vision summer school. In 2021, I was an invited participant in DOE’s AI for Earth System Predictability Workshop’s “Explainable/ Interpretable/ Trustworthy AI Session”.

I served as a program committee member for AAAI/ACM AIES conference in 2025 and 2026. I was a research associate in AI Governance at the Future Impact Group, working on Open Source AI Governance. I also served as the Secretary at IEEE Standards Association for standards-setting around measuring the energy use of AI systems. From 2024-2025, I was selected to join the Science Policy cohort (2024-2025) for AGU’s Voices for Science program.

Prior to this I was a NASA Postdoctoral Fellow at Goddard Space Flight Center and USRA, working with the Black Marble Science Team. I received a PhD in Computer Engineering with Profs. Antonia Papandreou-Suppappola and Philip Christensen at Arizona State University in 2019 focusing on “unsupervised modeling of satellite image time-series for change and novelty detection in Earth and Planetary observations“. While in grad school, I was a ML intern in NASA JPL and Los Alamos National Laboratory.

Broadly, I am interested in advances in frontier AI, its beneficial use cases such as those in scientific discovery, and at the interface between frontier AI and policy to safety develop the technology.

Contact: srija[dot]c[dot]chakraborty[at]gmail[dot]com