Saiteja Samudrala optimized the Titan Training Framework in the huggingface/torchtitan repository by migrating its training workflow for LLAMA3 8B to a DCP ZOC-based approach. Using Python and leveraging deep learning and distributed systems expertise, Saiteja replaced the default asynchronous and pinned memory model to improve training efficiency and resource stability. The work included enhancing checkpoint management and strengthening asynchronous operations, which streamlined workflows and reduced wait times. All modifications were delivered as a single, auditable commit, ensuring traceability. This focused engineering effort addressed workflow reliability and performance, demonstrating depth in both machine learning infrastructure and distributed training optimization.