Worked on the apache/parquet-java repository to deliver dictionary reuse support in Parquet’s Java implementation. Introduced a decode method to DictionaryPage and refactored ColumnReaderBase to leverage this new path, simplifying dictionary initialization and improving code clarity. Developed the testDeduplicatedDecodedDictionary to verify correct behavior and ensure robust decoded dictionary reuse. These changes targeted enhanced decoding performance and memory efficiency for large-scale Parquet workloads, while also increasing test coverage. The work was accomplished using Java and data engineering skills, with a focus on optimizing Parquet’s internal handling of dictionary pages. No major bugs were reported during this period, reflecting careful implementation.