Title: ORAQL: Orchestration of AI for Quantum Leadership
Abstract: Deciding whether a nuclear Hamiltonian is better simulated classically or on a quantum computer still relies on manual, case-by-case workflows. We present an agentic framework in which a large language model orchestrates validated solvers. The framework takes shell-model Hamiltonians from BIGSTICK, maps them to qubits (Jordan–Wigner, Bravyi–Kitaev, and Gray code), and produces auditable estimates of the qubits and gates required by VQE, ADAPT-VQE, and phase estimation. We discuss the trade-off between fewer qubits and denser Hamiltonians, and the path to medium-mass nuclei.