Lightbend's Karl Wehden will look at the typical decisions you will face as you mature your machine learning enhanced applications and start delivering at scale, on never ending streaming data. What trade-offs are you making when selecting a scoring strategy? What model metadata should you capture? How do you ensure model management and training is scalable and reliable? What should your serving and scoring infrastructure look like? These topics and more in a comprehensive view of how Lightbend and Red Hat are bringing repeatable, reliable, open source machine learning infrastructure to the market.