This poster presents the NSF-funded EnviStor, a smart data infrastructure that integrates storage, computation, metadata, and visualization tools to support environmental and climate data research. It highlights the implementation of a Smart Data Pipeline using Agentic AI to streamline data discovery, harmonization, cleaning, and visualization within high-performance computing environments. The project explores the use of AI agents in managing large, heterogeneous datasets and improving metadata generation and discoverability, while enabling collaboration through Open Science networks. Future directions include evaluating the accuracy of AI-assisted tasks across varied environmental data use cases.
