Open Source MLOps framework MLFlow launched by Databricks has been moved to the Linux Foundation. The news was announced at the Spark+AI Summit by Matei Zaharia, creator of Apache Spark and MLFLow, and the co-founder of Databricks. New functionalities are added to the framework, including integration with Continuous Integration/Continuous Deployment (CI/CD) platforms like GitLab and Jenkins, new Application Programming Interfaces (API), automatic versioning and logging, etc.
Over 100,000 Machine Learning models are added to the model registry every week and nearly 1 million experiments run on the platform. The shift was done to make the platform the vendor-neutral, joining the likes of Cloud Native Computing Founding, Kubernetes, DeltaLake, etc.