Worked on the NVIDIA/spark-rapids repository to enhance metrics readability for large-scale data processing workloads. Developed and integrated new formatting utilities in Scala to support petabyte and exabyte units, ensuring that data sizes exceeding one terabyte are displayed with accurate and consistent labeling across dashboards. This improvement addressed the challenge of interpreting very large data metrics, supporting better observability and capacity planning for backend systems. The work focused on backend development and data processing, strengthening the repository’s metric display logic and paving the way for future enhancements. No bugs were fixed during this period, with efforts concentrated on feature delivery.