Building an ML Stack for the Enterprise

Machine learning is undoubtedly an important factor to driving business value, but why do most organizations struggle so greatly with getting those valuable machine learning models into production? The answer seems to be that most machine learning teams follow standard software engineering practices, and although similar, there are strategies and tools that differentiate a machine learning model from a standard software lifecycle. From this issue the term MLOps was birthed, as well as a decision framework that guides enterprises to employ the tools that work best for their business. This session will introduce this decision framework, in addition to what components makes a superior ML stack. Finally, participants will get a sneak peak into a live demo that utilizes several of the tools to create a seamless machine learning workflow.