During April 2025, Marco Piannucci focused on improving metadata robustness in the zarr-python repository. He addressed a nuanced issue in Zarr Format 3 by fixing the encoding of special floating-point values such as NaN and Infinity during consolidated metadata serialization. Using Python and RST, Marco implemented a helper function that converts these values to string representations, ensuring consistency with ArrayMetadata behavior. This targeted bug fix enhanced cross-compatibility and reliability of metadata handling across the Zarr ecosystem. His work demonstrated depth in data serialization, metadata handling, and testing, resulting in more robust and accurate metadata encoding for Zarr users.