Mathematical modeling is an essential step, for example, to analyze the\ntransient behavior of a dynamical process and to perform engineering studies\nsuch as optimization and control. With the help of first-principles and expert\nknowledge, a dynamic model can be built, but for complex dynamic processes,\nappearing, e.g., in biology, chemical plants, neuroscience, financial markets,\nthis often remains an onerous task. Hence, data-driven modeling of the dynamics\nprocess becomes an attractive choice and is supported by the rapid advancement\nin sensor and measurement technology. A data-driven approach, namely operator\ninference framework, models a dynamic process, where a particular structure of\nthe nonlinear term is assumed. In this work, we suggest combining the operator\ninference with certain deep neural network approaches to infer the unknown\nnonlinear dynamics of the system. The approach uses recent advancements in deep\nlearning and possible prior knowledge of the process if possible. We also\nbriefly discuss several extensions and advantages of the proposed methodology.\nWe demonstrate that the proposed methodology accomplishes the desired tasks for\ndynamics processes encountered in neural dynamics and the glycolytic\noscillator.\n