Talk: DevOps for Data Science

We often start our data science discussions assuming data is ready to analyse and able to give us a predictive model that we can rely on. Little thought is given to practices enabling us to continually add value.

Modern software development is agile, but more and more our software relies on predictive models whose development is decidedly not agile. Often overlooked in data science discussions are documentation, creation and refresh of data and models, testing, monitoring model performance, and even ensuring data security and compliance.

In this session, see how DevOps best practices can be applied to Data Science to meet all these requirements. I'll walk through the practices, and the tools to enable you to get the most out of your Data Science efforts.