From Kubernetes to Kubeflow
What is Kubeflow?
Kubeflow as a project got its start over at Google. The idea was to create a simpler way to run TensorFlow jobs on Kubernetes. So, Kubeflow was created as a way to run TensorFlow, based on a pipeline called TensorFlow Extended and then ultimately extended to support multiple architectures and multiple clouds so it could be used as a framework to run entire machine learning pipelines. The Kubeflow open source project (licensed Apache 2.0) was formally announced at the end of 2017.
In a nutshell, Kubeflow is the machine learning toolkit that runs on top of Kubernetes. Kubeflow’s combined components allow both data scientists and DevOps to manage data, train models, tune and serve them, as well as monitor them.
About the Instructor
Chase is a professional cloud native geek with a passion for connecting business problems to engineering solutions. Chase is currently investing his time in helping clients get the most out of their MLOps platforms (more specifically Kubeflow) by improving feedback loops and tightening up their model development life cycle.