AI Seminar: Designing AI models for impact, the case of Clay, from idea to deployment
Join us for a talk by Bruno Sánchez-Andrade Nuño,Executive Director of Clay, the AI for Earth nonprofit. Everybody is welcome to attend.
Title
Designing AI models for impact, the case of Clay, from idea to deployment
Abstract
In this talk, we present Clay, our non-profit AI organization, as a case study in addressing the challenges of operational AI for Earth. We will outline our decision-making framework, from the initial ideation phase—considering why to start such an initiative, the choice of organizational structure, and our primary objectives—to the technical aspects of data selection, including the rationale behind choosing specific instruments. Additionally, we will delve into our architectural choices, such as the use of visual transformers and masked autoencoders, and the specific modifications made to ensure the system is agnostic to Earth observation instruments and resolutions. Finally, we will share our operational results, highlighting that developing the model is only the beginning of the journey.
Bio
Bruno is the Executive Director of Clay, the AI for Earth nonprofit, where he focuses on making AI learn Earth with open-source, open-data and open for business. He co-founded and leads the project. Previously, Bruno led the creation of the Planetary Computer at Microsoft. His past roles also include leading data projects at the World Bank and Chief Scientist at Mapbox, emphasizing his extensive experience in technology, and environmental and social impact. Bruno's book "Impact Science: The science of getting to radical social and environmental breakthroughs" highlights his commitment to science for social good. Recognized as a Young Global Leader by the World Economic Forum, Bruno's contributions to science and policy have been widely acknowledged. Originally from Asturias, Spain, Bruno currently resides in Copenhagen with his family.
This seminar is a part of the AI Seminar Series organised by SCIENCE AI Centre. The series highlights advances and challenges in research within Machine Learning, Data Science, and AI. Like the AI Centre itself, the seminar series has a broad scope, covering both new methodological contributions, ground-breaking applications, and impacts on society.