Open Source AI At Red Hat: Our Journey In The Kubeflow Community
Red Hat News, Tuesday, March 19th, 2024
Red Hat has been involved in AI and data projects for many years, and while many support the modernization journey, none tell the whole user journey to bring artificial intelligence (AI) and machine learning (ML) to production. This is why we launched Open Data Hub-to help develop and support open source projects dedicated to data and AI/ML, like Kubeflow and KServe.
Open Data Hub and Kubeflow
Open Data Hub started with a simple operator that deploys AI/ML and data software on Red Hat OpenShift clusters with some level of integration between them. However, when we began to hit roadblocks when looking for supportability, we turned to Kubeflow-a new project focused on simplifying and scaling ML workload deployments on Kubernetes.
By rewriting Open Data Hub on top of Kubeflow code, we were able to create a new support matrix of features. Red Hat has increased our involvement with the Kubeflow community over the past year to not only contribute code, but to assist the community in achieving higher levels of maturity and user acceptance. Below is an overview of the work we are currently engaged in the Kubeflow community.