Physics informed machine learning has been emerged as a powerful tool with the help of deep learning as the latter has been instrumental as a data-driven function approximator. Many recent works have been focusing on solving hard to solve differential equations with the help of physics informed neural network (PINN), a tremendously simple approach which blends physics and deep learning. We explore the application of PINN in solving Kuramoto system of coupled differential equations as well as in decision making problem of synchronization state of the system. The experimental results illustrate that PINN can not only be used to solve the coupled differential equations but be applied to figure out the synchronization capability of the oscillator system in consideration.