This project implemented a Deep Learning-based Email Spam Detection pipeline using a Bidirectional LSTM architecture trained on combined SpamAssassin and Enron datasets. Using sequence tokenization and a trainable embedding, the Bi-LSTM model achieved strong classification performance (insert your exact metrics from the run).