Oga T. developed the Upsample Interpolation Mode Extension for the ml-explore/mlx repository, focusing on enhancing the Upsample layer by adding cubic interpolation support. Using Python and leveraging skills in library development and type hinting, Oga updated the API to allow finer control over interpolation methods, which improves the flexibility and quality of downstream machine learning pipelines. The work emphasized type safety and clear API design, ensuring the feature aligns with repository standards and is ready for broader adoption. Although no bugs were fixed during this period, the contribution addressed a specific need for more robust upsampling options in ML workflows.