This study presents an innovative path through the method of deep learning to inspect the mixed shock response of a latitudinally-graded agglomerated carbon-nanotubes enriched composite (LaG-ACNTEC) rectangular plate. The case study of this research is simultaneously exposed to thermomechanical and heat flux wave shocks at its edge expanded through the latitudinal direction. Equations of motion for this model are established on the bases of the linear thermoelasticity theory. Differential quadrature technique (DQT) is utilized for solving spatial dependency of these equations, while the Laplace transform and its inverse are utilized to solve temporal dependency of the system’s equations. Authors of this study reveal a novel path for increasing the computation efficiency of the approach by mounting a deep-learning-based method on the results determined by DQT and Laplace transform at designated points due to dynamic nature of the problem. Verification of the applied solution is performed based on a comparative study with the results of the published surveys. This study gives precious insights to designers for inspecting the mixed shock response of the composite structures by considering the effect of agglomeration and utilizing data-driven methods serving as accelerator of conventional solvers.