Developed MapType casting support in Spark for the apache/datafusion-comet repository, enabling seamless casting between MapType structures and integrating these operations with Spark’s native cast_map_to_map functionality. Focused on robust data engineering practices, the work included implementing comprehensive tests to validate both compatible and incompatible casting scenarios, ensuring reliable propagation of data type compatibility. By leveraging Scala and Spark, the solution reduced the need for manual data transformations within data pipelines and improved overall interoperability between Spark and datafusion-comet. No major bugs were addressed during this period, with efforts concentrated on feature development and thorough validation of new casting capabilities.