Deployed ML models need your feedback too

MLOps has taken a giant leap forward in the last few years, but actionable monitoring and management of AI models in production remains a struggle. Only recently, large cloud providers and major open-source projects have started to provide the means to perform tasks like data drift detection or creating feedback loops. How far can you take model monitoring without feedback? What is the current state-of-the-art? What is missing to close the MLOps circle? These questions will be addressed, and a way to provide feedback to cloud AI services will be demonstrated.