Describe your approach to setting up a CI/CD pipeline for a machine learning project at Berkeley Research Group.
