Deploying Machine Learning Models Successfully into Production
Field CTO for EMEA, Cloudera
One question that keeps coming up when working with our customers is: What are our recommendations for best practices in implementing machine learning? The general response is, unfortunately – it depends. Machine learning operations (MLOps) is considered the next frontier for deploying fully automated, fully operational ML products and models that magically manage, monitor and update themselves.
However there is no standard or even a widely adopted framework for this and most enterprises are coming up with something that fits what they already have. In this talk I will go through the most common machine learning workflows that I see being implemented in the wild, what works, what doesn’t and why. The talk will also cover some of the new continuous development approaches to machine learning and how they differ from the software development equivalents.
Dr Chris Royles is Field CTO for EMEA at Cloudera. He helps organisations innovate through the use of data, working across industries that are regulated and organisations where data privacy is critical. Chris‘s focus is on the development of skills and methods for migration to the Enterprise Data Cloud. Chris holds a Ph.D. in Artificial Intelligence from Liverpool University.