We introduce a classical computational method for quantum dynamics that relies on a globalin-time variational principle. Unlike conventional time-stepping approaches, our scheme computes the entire state trajectory over a finite time window by minimizing a physically motivated loss function that enforces the Schrödinger's equation. The variational state is parametrized with a Galerkin-inspired ansatz based on a time-dependent linear combination of time-independent Neural Quantum States. This structure effectively captures the relevant dynamical frequencies in the time evolution and is particularly well-suited for exploring long-time dynamics. We showcase the method's effectiveness by simulating global quantum quenches in the paradigmatic Transverse-Field Ising model in both 1D and 2D. By extracting the asymptotic long-time evolution, we uncover signatures of ergodicity breaking and absence of thermalization in two dimensions. Overall, the method presented here shows competitive performance compared to state-of-the-art time-dependent variational approaches and highlight the potential to explore previously inaccessible dynamical regimes of strongly interacting quantum systems.
