Worked on the awslabs/graphstorm repository to enhance dataset handling and streamline local development workflows. Focused on configuration management and data handling, the work introduced normalization for dataset names, allowing both 'ogbn-papers100M' and 'ogbn-papers100m' to be recognized seamlessly across environments. Using Python and Shell scripting, implemented a .gitignore update to exclude the dataset directory, preventing accidental tracking of generated files and improving repository hygiene. Additionally, addressed a common command typo to support reproducibility and reduce onboarding friction. These changes contributed to more robust data workflows, minimized environment-specific discrepancies, and supported faster, cleaner setup for new developers.