Frictionless AI development using Docker containers

Making the right architectural decisions about how to house your AI classifiers can make a world of difference at all levels of your engineering workflow. At StoryStream we use Docker containers to ease each and every stage of delivering classifiers to our end users.  From being able to develop and test new algorithms, through to staging and scaling out to a cloud-based production environment. Docker can make your life as an AI developer easier, more efficient and more enjoyable. In this talk I’ll show how you can use Docker containers to make your classifiers completely portable, fit into all your existing development workflows (source control, CI/CD), and discuss various options on how to host your classifiers on AWS.

Required Audience Experience

No prior experience with Docker. This session will discuss focus on high level concepts instead of technical details as much as possible.

Track 2
Location: Gielgud Date: September 30, 2019 Time: 5:05 pm - 5:50 pm Joe Baskerville, StoryStream