Worked on enhancing the Row Tracking subsystem in the apache/paimon repository, focusing on improving performance and reliability. Introduced a fail-fast assertion in the Python code path to prevent conflicts when file_source is missing, and refined the logic for handling _ROW_ID to avoid duplicate entries during row-tracking operations. Added targeted test coverage to verify filter pushdown on the row-tracking system table, which accelerates data retrieval. The work spanned both Python and Spark components, aligning cross-component behavior for end-to-end robustness. Leveraged skills in database management, backend development, and Scala to deliver a feature that strengthens data integrity and system efficiency.