Sep 9–12, 2019

Getting started with Kubeflow

Skyler Thomas (MapR)
9:00am12:30pm Tuesday, September 10, 2019

Who is this presentation for?

Data Scientists and AI Infrastructure Engineers



Prerequisite knowledge

Knowledge of Kubernetes and various machine learning frameworks like Tensorflow and Pytorch would be extremely useful but is not required. The ability to clone a GitHub repo is crucial

Materials or downloads needed in advance

Attendees will need a laptop with connectivity to GitHub and a >v1.11 Kubernetes environment. This Kubernetes environment can be a minikube installation on your laptop but you will get the most out of the course if you have a Kubernetes cluster with several nodes available.

What you'll learn

Attendees will understand the architecture of Kubeflow. They will be able to install Kubeflow into Kubernetes. They will be able to use Jupyter Notebooks, start a Tensorflow training job, and serve their models to handle requests.


Kubeflow consists of dozens of components used to train and serve machine learning models. Our first task will be to understand the Kubeflow architecture and how it fits into the larger Kubernetes ecosystem. We will discuss in detail the performance, availability, and security impacts of various Kubeflow deployment options and decisions that you might make in how you use Kubeflow.

Next, we will go hands on and actually deploy a subset of the Kubeflow components into a running Kubernetes environment. We will then use Jupyter Hub and Jupyter notebooks to import our training data and set up prerequisites for creating training environments using Tensorflow (and PyTorch time permitting) Finally, we will being to submit training jobs and generate models. We will serve these models via Tensorflow Serving (and Seldon Core time permitting)

Photo of Skyler Thomas

Skyler Thomas


Skyler Thomas is an engineer at MapR, where he is designing Kubernetes-based infrastructure to deliver machine learning and big data applications at scale. Previously, Skyler was an architect at IBM, where he worked with more than a hundred customers to deliver extreme-scale applications in the healthcare, financial services, and retail industries.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)

Contact us

For conference registration information and customer service

For more information on community discounts and trade opportunities with O’Reilly conferences

For information on exhibiting or sponsoring a conference

Contact list

View a complete list of O'Reilly AI contacts