Fueling innovative software
July 15-18, 2019
Portland, OR

Introduction to Kubeflow Pipelines

Dan Anghel (Google)
1:30pm5:00pm Tuesday, July 16, 2019
Secondary topics:  AI Enhanced
Average rating: ****.
(4.50, 6 ratings)

Who is this presentation for?

  • Data scientists, data engineers, and developers interested in ML




The Kubeflow project is dedicated to making deployments of ML workflows on Kubernetes simple, portable, and scalable.

Dan Anghel dives into installing and using Kubeflow Pipelines to create a full machine learning application on Kubernetes so you can become familiar with Google Cloud Platform tools such as Cloud Shell and Kubernetes Engine. You’ll start with an empty environment and create a Kubernetes cluster and install Kubeflow from scratch. You can build and run a full pipeline that does distributed training of a TensorFlow model, then scales and serves the trained model and deploys a web frontend for requesting predictions from the model. Dan teaches you how to use a Jupyter notebook to build and run a pipeline using the Kubeflow Pipelines SDK.

Prerequisite knowledge

  • A basic knowledge of Kubernetes (useful but not required)

Materials or downloads needed in advance

  • A laptop with a modern browser such as Chrome installed

What you'll learn

  • Learn how to install and use Kubeflow Pipelines to support ML workflows on Kubernetes and use the Pipelines SDK from a Jupyter notebook
Photo of Dan Anghel

Dan Anghel


Dan Anghel is a strategic cloud engineer with Google after a more than 10 years’ long adventure in retail. Specialized in machine learning and big data, he’s helping the largest Google customers accelerate their journey into the cloud. Besides AI and machine learning, he’s been passionate about metal music since childhood, so there’s a great chance you will find him at the concert whenever a cool band comes to town.