Back

Speaker "Josh Bottum" Details Back

 

Topic

Kubeflow workshop

Abstract

As demonstrated by the thousands of attendees in the (14) Kubeflow sessions at KubeCon San Diego, Kubeflow continues to be the leading Kubernetes-based, open source project for machine learning. This session targets data scientists, ML engineers and product managers, who want to leverage Kubernetes to scale up their Machine Learning experiments.    In this presentation, Mr. Bottum will review the Kubeflow Community and examine ML use cases from market leaders.  Additionally, Mr. Bottum will discuss the features in the latest Kubeflow release, Kubeflow 1.0 and provide a quick Kubeflow workflow demonstration.

Attendees will learn a) the basics of Kubeflow, the ML toolkit for K8s, and b) how to build and deploy complex data science pipelines with Kubeflow Pipelines.  The tutorial review will focus on two essential aspects: 1. Low barrier to entry: deploy a Jupyter Notebook to Kubeflow Pipelines using a fully GUI-based approach. This workflow enables data scientists to exploit the scaling potential of K8s - no CLI commands, SDKs, or K8s knowledge required. 2. Reproducibility: automatic data versioning and volume snapshots will enable full reproducibility and collaborative development, as well as fine grained analysis and visualizations after pipeline executions. 


 

Profile

Josh Bottum is a Kubeflow Community Product Manager.  His Community responsibilities include assisting users to quantify Kubeflow business value, develop critical user journeys (CUJs), triage incoming user issues, prioritize feature delivery, write release announcements and deliver presentations and demonstrations of Kubeflow. Mr. Bottum is also a VP of Arrikto. Arrikto simplifies storage operations for stateful Kubernetes applications by enabling efficient local storage architectures with data durability and portability. Arrikto is a core code contributor to Kubeflow.