This document discusses using recurrent neural networks and auto-regressive recurrent neural networks to predict trends in COVID-19 cases in India. It first provides background on the COVID-19 pandemic and reviews several previous studies that have used machine learning techniques like LSTM, GRU, and Bi-LSTM to predict COVID-19 cases. The document then describes implementing RNNs like LSTM and GRU independently as well as using deepAR with LSTM and GRU cells on a COVID-19 dataset from India to make predictions. The mean absolute percentage error and root mean square error metrics will be used to evaluate the results.