<p>Many scholars are still debating the idea of employability, and no agreement has been reached as of yet. A solid theoretical foundation is still not established with the empirical investigation. The ability of the graduate to find fulfilling employment emphasizes the satisfaction that the graduate finds in their job. Therefore, employability prediction models are crucial in assessing a student's ability to find employment. This study aims to create a hybrid optimization-enhanced deep learning model for predicting employability. For that, primarily, input data is given to the pre-processing phase, where quantile normalization is used. Then, feature fusion is done by using Renyi entropy with Generative Adversarial Network. The employability prediction is done by utilizing a Deep Recurrent Neural Network (Deep RNN) in which its weight is trained using the proposed Chronological Squirrel Search Algorithm (CSSA). Here, CSSA is constructed by the combination of the Chronological concept and Squirrel search algorithm to optimize the predicted result. Moreover, the predicted output is noted. Furthermore, the introduced Chronological Squirrel Search Algorithm_Deep Recurrent Neural Network (CSSA_Deep RNN) compared with different algorithms illustrates better performance concerning the evaluation metrics such as Root-Mean-Square Error and Mean Square Error with a minimal error value of 0.458 and 0.210, respectively.</p>