TensorFlow London: Training to Explainability via GitOps with Kubeflow by Ryan Dawson
Speaker: Ryan Dawson, Cloud Engineer at Seldon
Title: Training to Explainability via GitOps with Kubeflow
Bio: Ryan Dawson is a core member of the seldon open source team, providing tooling for machine learning deployments to Kubernetes (https://github.com/SeldonIO/seldon-co.... He has spent 10 years working in the Java Development scene in London across a variety of industries.
Abstract: Ryan will show models being trained and deployed for serving from kubeflow pipelines via gitops, highlighting git-powered audit trails and then walk through some of the observability features using Seldon with kubeflow, focusing on grafana-based monitoring. He will use the EFK stack for request logging and then perform explanations on the requests using the open-source library alibi. All of this behind the kubeflow gateway's istio authentication.