Worked on enhancing data manifest handling in the trinodb/trino repository by updating the ManifestFiles.read method to accept specsById, enabling more reliable and flexible processing of data manifests. This change required careful backend development in Java, with a focus on improving the robustness of manifest operations. Unit tests were updated to accurately reflect the new specsById-driven workflow, ensuring that the enhancements were thoroughly validated and integrated into the existing codebase. The work emphasized maintainability and correctness, contributing to better test coverage and more dependable data management within the project, while leveraging strong skills in Java and backend engineering practices.