Developed and integrated the Independent Subnet Training (IST) built-in function for Apache SystemDS, enabling distributed neural network training across independent subnets to enhance parallelism and resource efficiency. This feature, implemented using DML and Java, established a foundation for scalable model training and future distributed training capabilities within the repository. The work focused on distributed systems design and built-in function development, with attention to end-to-end code contribution and integration. No major bugs were addressed during this period, as efforts centered on delivering this foundational feature and ensuring code quality, unlocking faster experimentation and improved utilization of computational resources for machine learning workflows.