This study addresses critical challenges in accurately predicting the Remaining Useful Life (RUL) of Lithium-ion batteries, a task vital for ensuring reliability and safety in various applications. Despite extensive research utilizing deep learning methods, existing approaches often face limitations in prediction accuracy and generalizability. To address these gaps, we propose a novel hybrid deep learning framework that integrates Deep Neural Networks (DNN) and Gated Recurrent Units (GRU) for improved RUL prediction. Our method establishes a new benchmark, demonstrating superior accuracy compared to prior methodologies, particularly those leveraging datasets from CALCE. Comprehensive performance evaluation, using metrics such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), highlights the exceptional predictive capability and robustness of our approach. This work contributes a significant advancement in RUL prediction, offering a scalable and reliable solution to a critical problem in battery health management.