Researchers at the Johns Hopkins Applied Physics Laboratory are developing agentic AI to enhance autonomy, coordination, and adaptability in collaborative robotic teams. A key challenge is designing a scalable architecture that supports agentic behaviors across heterogeneous systems. To address this, the laboratory has introduced a novel approach that leverages large language models (LLMs) to create AI agents capable of complex decision-making. This architecture enables robots to learn from experience and adapt to new situations, facilitating more effective teamwork. The laboratory's research has yielded valuable insights into the development of agentic AI, including the importance of scalable design and the potential applications of LLM-based AI agents1. This work has significant implications for the field of robotics, as it could enable more sophisticated and autonomous robotic systems, so what matters to practitioners is how these advancements could be applied to real-world scenarios, potentially transforming industries such as manufacturing and logistics.