This graduate-level course explores the cutting-edge intersection of large language models (LLMs) and robotic control systems, focusing on how AI agents can autonomously execute real-world robotic tasks. Students will gain hands-on experience designing, implementing, and evaluating agentic AI systems that bridge the gap between natural language understanding and physical robot manipulation. Through a combination of theoretical foundations and practical experimentation, students will work with physical robotic arm kits equipped with IMU sensors, learning to develop AI agents using industry-standard frameworks (Microsoft AgentFramework). The course emphasizes rigorous evaluation methodologies, teaching students to assess AI performance using precision, recall, F1 scores, and statistical significance testing.