My name is Alex Aidun and I joined Arrikto on August 30th, 2021 as the Director, Education to build upon the great work done by Shannon Bradshaw of Unusual Ventures. I have spent the majority of my career focused on education and documentation for end users of enterprise solutions. During this time I’ve had the privilege of working with amazing educators and learning leaders while being able to use and develop an appreciation for cloud technologies. I am thrilled to join Arrikto and focus on educating Kubeflow users to further our mission of evangelizing and facilitating the MLOps potential of Kubeflow through the enablement of data scientists. This is a truly exciting time for this community and we are looking forward to partnering with everyone to provide the best education we can.
A Skills Based Approach to Kubeflow Education
Kubeflow demands a skills-based approach to education, a methodology with proven effectiveness. When we consider how to enable data scientists to be successful we consider the Bloom’s Taxonomy Pyramid.
This diagram clearly illustrates the different tiers of understanding audiences may have with technical concepts. Our goal at Arrikto is to facilitate a higher level of comprehension than remember or understand. We are focused on enabling data scientists to analyze and transform existing Python code to create Kubeflow pipelines and to evaluate machine learning models for quality and ultimately deploy inferencing services to production. We want to teach data scientists and engineers to be self-sufficient with Kubeflow. Viewing education through this lens ensures that we are building truly useful and first-in-class education for our audiences. This also ensures that audiences experience a clear skills progression as they navigate through learning paths and education programming.
Kale 101 and Katib 101 Courses are Live!
We’ve recently released our first two courses – Kale 101 and Katib 101 – both of which focus on the necessary Day 1 Fundamentals that all data scientists need to be successful with the Kubeflow ecosystem. We know that learning is a joint effort – our audiences are going to give us time and attention and in return expect quality programming that is engaging and informational. All our courses will be fundamentally hands-on. Learners will work through key concepts and recommended practices in the MiniKF Kubeflow environment. We’ll deliver instruction through video tutorial based concept discussion supplemented by additional reading for those that want deeper dives. The courses are capped off with self-graded labs to ensure skills are being developed and retained.
Kubeflow Education is a Community Effort
We are working aggressively to build out more Kubeflow education on Arrikto Academy and we would like to invite you along on our journey. For now we are using MKDocs as our static site builder solution, however sometime in 2022 we will transition to a more formal Learning Management System. Upcoming curriculum will focus on:
- Snapshotting with Rok & Marshal Volumes
- Serving Models via Kale, KFServing and Inference Servers
- Version Control and Management
- Kubeflow Pipeline Analysis and Troubleshooting
The repo is publicly available and we encourage you to PR issues or requests or fork and try your hand at building content!
What’s Next
- Advance your Kubeflow education – head over to our Academy portal, try the two new courses and give us some feedback.
- We are updating documentation to be feature focused in addition to providing procedural and step by step support.
- Staying connected via LinkedIn, Twitter or community Slack will ensure you are most up to date with our progress.
- Check out our fundamentals sessions and associated webinars as well as our meetups.
About Arrikto
At Arrikto, we are active members of the Kubeflow community having made significant contributions to the latest 1.4 release. Our projects/products include:
- Kubeflow as a Service is the easiest way to get started with Kubeflow in minutes! It comes with a Free 7-day trial (no credit card required).
- Enterprise Kubeflow (EKF) is a complete machine learning operations platform that simplifies, accelerates, and secures the machine learning model development life cycle with Kubeflow.
- Rok is a data management solution for Kubeflow. Rok’s built-in Kubeflow integration simplifies operations and increases performance, while enabling data versioning, packaging, and secure sharing across teams and cloud boundaries.
- Kale, a workflow tool for Kubeflow, which orchestrates all of Kubeflow’s components seamlessly.