Across the ML lifecycle, there a number of tools and strategies to streamline the operations for ML development teams. 1. Creating data stores and feature stores for teams that need to access similar datasets/model artifacts 2. Created a model and container registry for teams to access containers/models for training/experimentation 3. Using open source frameworks for model training (HuggingFace, Pytorch) 4. Using cloud infrastructure for model inference (using terraform/CDK to manage that infrastructure)) 5. Using shared code repositories to share information/scripts between related teams (Gitlab)