Machine Learning Design Patterns for MLOps // Valliappa Lakshmanan // MLOps Meetup #49

MLOps community meetup #49! Last Wednesday we talked to Lak Lakshmanan, Data Analytics and AI Solutions, Google Cloud.
// Abstract:
Design patterns are formalized best practices to solve common problems when designing a software system. As machine learning moves from being a research discipline to a software one, it is useful to catalog tried-and-proven methods to help engineers tackle frequently occurring problems that crop up during the ML process. In this talk, I will cover five patterns (Workflow Pipelines, Transform, Multimodal Input, Feature Store, Cascade) that are useful in the context of adding flexibility, resilience, and reproducibility to ML in production. For data scientists and ML engineers, these patterns provide a way to apply hard-won knowledge from hundreds of ML experts to your own projects.
Anyone designing infrastructure for machine learning will have to be able to provide easy ways for the data engineers, data scientists, and ML engineers to implement these, and other, design patterns.
// Bio:
Lak is the Director for Data Analytics and AI Solutions on Google Cloud. His team builds software solutions for business problems using Google Cloud's data analytics and machine learning products. He founded Google's Advanced Solutions Lab ML Immersion program and is the author of three O'Reilly books and several Coursera courses. Before Google, Lak was a Director of Data Science at Climate Corporation and a Research Scientist at NOAA.
---------- ✌️Connect With Us ✌️------------
Join our Slack community: go.mlops.community/slack
Follow us on Twitter: @mlopscommunity
Sign up for the next meetup: go.mlops.community/register
Catch all episodes, Feature Store, Machine Learning Monitoring and Blogs: mlops.community/
Connect with Demetrios on LinkedIn: / dpbrinkm
Connect with Lak on LinkedIn: / valliappalakshmanan
Timestamps:
[00:00] TWIML Con Debate announcement to be hosted by Demetrios on Friday
[00:19] Should data scientists know Kubernetes?
[05:50] Lak's background in tech
[08:07] Which ones you wrote in the book? Is the airport scenario yours?
[09:25] Did you write ML Maturity Level from Google?
[12:34] How do you know when to bring on perplexity for the sake of making things easier?
[16:06] What are some of the best practices that you've seen being used in tooling?
[20:09] How did you come up with writing the book?
[20:59] How did we decide that these are the patterns that we need to put in the book?
[22:20] Similar problems in all verticles
[24:14] The "audacity" to think this is something that is worth doing?
[26:10] "There's a need to explain these concepts and patterns and that's the genesis of the book."
[31:29] Hierarchy of design patterns?
[32:01] Hierarchy or handcuff?
[32:39] Treat quality in a statistical way."
[38:05] Are there patterns yet to be discovered?
[38:43] "There are patterns out there that we did not include in the book
[42:08] ModelOps vs MLOps
[43:08] DevOps engineer transition to Machine Learning engineer
[43:36] "I haven't seen data scientists becoming ml engineers."
[46:07] Fundamental Machine Design Patterns vs Software Development Design Patterns
[49:23] DevOps is often mistaken as just a pure toolchain
[51:31] Advanced Solutions Lab
Advanced Solutions Lab:
cloud.google.com/asl
Purchase the book: Machine Learning Design Patterns: Solutions to Common Challenges in Data Preparation, Model Building, and MLOps here:
www.amazon.com/_/dp/109811578...

[@the-ghost-in-the-machine1108]

You're doing an excellent job facilitating inspiring conversations, mate. Keep up the great work!

[@roy11883]

Enjoyed this conversation, will definitely check out this book to get the patterns used in many customer use cases. Thank you MLOps team for bringing Lak to this show.

[@Baron-digit]

I've just started the GCP course on Coursera, and I have to say I really enjoy listening to this guy ;-)

[@MLOps]

yes he is super knowledgable!

[@amanditafbri]

What an amazing conversation!

[@MLOps]

Thank you Amanda! Lak is a real Gem to talk to so it was quite easy and flowing

[@IronOver]

I agree if you require a data scientist to know Kubernetes, you failed in designing your stack.

[@MLOps]

Totally!

[@krtst]

What about startups?

[@IronOver]

@@krtst startups are lucky they do not have legacy systems. I would choose one of the cloud provider solutions depending upon needs. A startup needs to focus on their core product and how that will be deployed for customers.