Developed a Deep Learning Script Suite for the se-sac/sesac-01 repository, enabling streamlined machine learning experimentation across MNIST, Fashion-MNIST, and transfer learning tasks. Leveraging Python, TensorFlow, and Keras, the work introduced reusable scripts for model definition, training, evaluation, and prediction, supporting both ANN and CNN architectures. The suite expanded to include transfer learning workflows for pneumonia detection and flower classification, incorporating data augmentation and image classification techniques. By establishing a robust directory structure and end-to-end ML pipelines, this contribution reduced friction in experimentation and accelerated feature development for AI-enabled products, laying a foundation for rapid prototyping and deployment.