Metaflow

ML and data science projects have a habit of becoming unmanageable as they grow — scattered code, hard-to-reproduce experiments, and infrastructure that doesn't scale. Metaflow is an open-source framework originally developed at Netflix for building and managing real-life ML, AI, and data science projects. It lets data scientists and ML engineers develop, debug, and deploy workflows in plain Python, with seamless access to cloud resources for scalable compute and data. Metaflow integrates with existing infrastructure, security, and data governance policies, and supports deployment on AWS, Azure, Google Cloud, and Kubernetes.