Agentic systems, in which diverse agents cooperate to tackle challenging problems, are exploding in popularity in the AI community. However, the agentic frameworks used to build these systems have not previously enabled use with research cyberinfrastructure. In this talk, I will introduce Academy, a modular and extensible middleware designed to deploy autonomous agents across the federated research ecosystem, including HPC systems, experimental facilities, and data repositories. I will situate Academy in the landscape of existing agentic frameworks and discuss how the design of Academy addresses the unique needs of scientific agents. Finally, I will present use-cases of agentic workflows and show an example of how we used Academy to build a more intelligent and efficient workflow for materials design.