Back

Speaker "Maria Vechtomova" Details Back

 

Topic

MLOPs as a product

Abstract

10 years ago, the first corporate companies started investing in data science. By now, most have multiple machine learning models running in production, but productionalizing new models takes longer than companies would like it to.
The goal of MLOps is to standardize the processes around machine learning model deployment and significantly reduce time to production. In large organizations, it quickly pays off.
I believe that MLOps should be viewed as a product and therefore managed as a product. In my talk, I will walk through the components of an MLOps platform (such as MLOps infrastructure, reusable deployment pipelines, and governance) and the composition of an MLOps product team.

Who is this presentation for?
AI/ data managers, data scientists, ML engineers
Prerequisite knowledge:
-
What you'll learn?
- MLOps is about people, not about tools; - All tools can be used in the wrong and in the right way, you need to know the MLOps principles

Profile

Maria is a Lead ML Engineer at Ahold Delhaize, bridging the gap between data scientists infra and IT teams at different brands and focusing on standardization of ML model deployments across all the brands of Ahold Delhaize. Maria believes that a model only starts living when it is in production. For this reason, last seven years, she focused on MLOps. Together with colleague Basak, started a blog https://medium.com/marvelous-mlops to share MLOps knowledge with other ML professionals.