Developed and delivered a feature for the harupy/mlflow repository that streamlines Databricks Model Serving by introducing automated environment packaging during model registration. Leveraging Python and Databricks, the work added an env_pack parameter to the register_model function, enabling dependencies and environment specifications to be bundled and version-controlled at the point of registration. This approach reduces deployment time and operational risk by ensuring all necessary packages are available when models are deployed on Databricks. The solution focused on MLOps best practices and model deployment reliability, with no major bugs reported during the development period, reflecting a focused and robust engineering effort.