Developed an end-to-end distributed training example for the pytorch-labs/monarch repository, focusing on demonstrating Monarch’s Distributed Data Parallel (DDP) workflow for convolutional neural networks on Oracle Cloud Infrastructure Kubernetes (OKE). The work centered on reproducibility and practical onboarding, providing detailed setup steps, configuration, and sample commands to help users replicate DDP training in production-like environments. Leveraging Python and YAML, the implementation showcased deep learning and distributed systems expertise, using a public dataset to illustrate the workflow. This contribution enhanced Monarch’s documentation and usability, enabling users to efficiently adopt distributed training practices within Kubernetes-based machine learning pipelines.