Worked on the brain-score/vision repository to address a data integrity issue in benchmarking workflows. Delivered a targeted bug fix in Python that caps benchmark scores at 1.0, specifically for the Lonnqvist2024 and Malania2007 benchmarks, using score normalization techniques to prevent inflated results. This adjustment ensures that evaluation metrics remain within valid bounds, supporting fair model comparisons and reliable reporting. The solution was implemented with careful adherence to project conventions and underwent code review before merging, reinforcing repository stability. Demonstrated skills in benchmarking and data normalization, contributing to the overall quality and consistency of model evaluation processes.