Dr. Diego Martinez is a data scientist at NASA’s Jet Propulsion Laboratory in Pasadena, California. He specializes in developing and deploying advanced machine learning systems for Earth and planetary science applications, with a focus on optimizing deep learning models for production environments and enabling rapid response to natural disasters. His work on the NASA JPL OPERA project involves optimizing self-supervised transformer models for emergency disaster response, enabling rapid damage assessment using Synthetic Aperture Radar data from Sentinel-1 satellites to detect earthquake, flood, fire, and landslide damage across diverse global environments.

Prior to his current role, Diego was instrumental in advancing Automated Machine Learning (AutoML) systems as part of DARPA’s Data-Driven Discovery of Models (D3M) program during his doctoral studies at Texas A&M University. He designed and developed high-performing AutoML systems, incorporating neural architecture search and Bayesian optimization techniques that significantly reduced user overhead while enhancing scalability.

Diego’s projects span multiple domains including developing semi-supervised learning frameworks for oceanographic data analysis with applications in eddy detection and heat flux reconstruction, implementing fault-tolerant visual-inertial odometry systems for Mars exploration vehicles, and creating comprehensive parallelization frameworks for atmospheric science experiments in Planetary Boundary Layer research. He is passionate about bridging the gap between cutting-edge machine learning research and practical deployment in space and Earth science applications.

Find my publications on Google Scholar

Contact

Education

• Computer Science (PhD) - Texas A&M University
• Computer Science (BS) - Autonomous University of Yucatan

Research Interests

• Earth Science
• Machine Learning
• Deep Learning
• Computer Vision
• Automated Machine Learning
• Large Language Models