Accelerating AI: Pillars and Pitfalls of ML Ops

Learn more about applying DevOps principles to accelerate the development and usage of machine learning models! In this talk, we will start with a definition of ML Ops and explore how DevOps principles can be applied to optimize machine learning through reproducibility, scalability and instrumentation. We’ll cover some of the challenges applying DevOps to machine learning models. We will explore some principles for building out an ML Ops toolset and pipelines tailored to your organization’s goals. Afterward, we will sketch out some ML Ops best practices, including how to build tracking, version control, and data quality monitoring into your ML Ops pipeline. Finally, we’ll touch on some common pitfalls and how to steer clear of them.



Christopher Pope

Christopher Pope is a seasoned DevOps / DevSecOps leader, with extensive experience enabling teams to build higher-quality, more secure solutions and deliver value to customers faster and more consistently.