Shivanjan Choudhury developed anisotropic voxel size support for volume allocation and NumPy data loading in the NVIDIA/warp repository, enabling users to specify non-uniform voxel spacing for scientific workloads. Using Python and leveraging skills in 3D graphics, CUDA programming, and numerical computing, Shivanjan introduced a centralized validator to normalize voxel sizes and extended API type hints to accommodate three-element voxel dimensions. The work included comprehensive unit testing for edge cases such as invalid inputs and NumPy scalars, ensuring robust validation across allocation paths. These enhancements improved data fidelity, reduced allocation errors, and streamlined analytics for non-uniform volumetric datasets.