Sea Surface Temperature (SST) is a key indicator of the global climate system. It is an essential factor in simulations of atmospheric models, weather predictions, and the study of marine ecosystems. Of these interests, one of the most important is studying changes in sea surface temperatures that result from the anthropogenic forcing of climate, a process known as global warming. An accurate prediction of sea surface temperature is highly beneficial towards understanding climate change, preserving marine ecosystems, etc. This research focuses on processing sea surface temperature data to perform a time series forecasting using a unique long short-term memory neural architecture to predict sea surface temperature in Bay of Bengal region. Given its reasonable high accuracy and low error rate, the assessment metrics show that the recommended neural network for SST prediction can be employed for real-time applications.