Akshita Bhagia contributed to the allenai/OLMo and OLMo-core repositories by delivering six features and resolving one bug over two months, focusing on backend development, CI/CD, and data engineering. She enhanced training data quality by introducing a new dataset mix and improved build reliability through upgrades to GitHub Actions workflows and dependency management using Python, TOML, and YAML. Akshita standardized training durations with configuration updates and aligned Python version policies to ensure compatibility and reproducibility. Her work addressed infrastructure efficiency, automation stability, and model training consistency, demonstrating a thoughtful approach to workflow management and machine learning operations in a collaborative environment.