AI-Enabled Optimization of Quantum Circuit Design for Realistic Nuclear Problems Genesis Mission Quantum computers have the potential to simulate nuclear systems beyond what can be described using classical computers. However, translating a nuclear physics problem into an executable quantum workflow requires navigating a vast design space. Researchers must choose among different methods for encoding the problem, select appropriate quantum algorithms, map those algorithms onto specific hardware architectures, and optimize the resulting quantum circuits. Each choice affects accuracy, computational cost and whether the calculation can run on available quantum processors. Making these decisions manually is time-consuming, requires specialized expertise across multiple domains, and often produces suboptimal results. An Argonne-led team is developing an artificial intelligence (AI) agent to automate this process. The system will learn to design quantum workflows by exploring the space of encoding schemes, algorithmic strategies and hardware configurations, then evaluating the resulting circuits against performance metrics such as gate count, circuit depth and expected accuracy. The AI agent will use reinforcement learning and other optimization techniques to identify efficient quantum workflows tailored to specific nuclear physics calculations. The project addresses a DOE Genesis Mission challenge area focused on discovering quantum algorithms with AI, specifically targeting quantum advantage for nuclear and hadronic systems. During the Phase I effort, the team will focus on representative nuclear structure and scattering problems, developing the AI framework and demonstrating its ability to generate competitive quantum circuits. The long-term goal is to create a generalizable tool that accelerates the adoption of quantum computing in nuclear physics, enabling researchers to tackle problems that remain inaccessible to both classical computation and manual quantum-circuit design.