This research aims to develop an automatic translation model based on Neural Machine Translation (NMT) for translation between Javanese and Indonesian. The method used is the Long Short-Term Memory (LSTM) model with the application of hyperparameter tuning and optimization using Adam's algorithm. This research collected data from Wulangreh and Wedatama, Javanese classical literature, and Indonesian translation data. The data is used to train and test the developed NMT LSTM model. However, the model performance evaluation results using the Bilingual Evaluation Understudy (BLEU) metric showed improved translation quality after applying hyperparameter tuning and Adam optimization. The developed NMT translation model has practical implications in facilitating communication between Javanese and Indonesian speakers, especially in document translation, social media, and e-learning platforms to preserve Javanese culture.