ModelOps refers to the process of deploying and updating analytical models on a regular basis by the data science team and the IT production team. ModelOps can be summarized as existing practices for putting machine learning models into production. It can be framed as an evolution/advancement to MLOps which focuses primarily on the automation of model training and feature engineering.
Gartner defines ModelOps (or AI operationalization) as the primary focus on the governance and life cycle management of a wide range of operationalized artificial intelligence and decision models, including machine learning, knowledge graphs, rules, optimization, linguistic and agent-based models. The main capabilities include continuous integration/continuous delivery, model development environments, champion-challenger testing, model versioning, model store and roll back.
Machine learning (ML) models are increasingly being used by organizations to turn massive amounts of data into valuable information. Data scientists are able to use these ML models to identify patterns in massive amounts of unstructured data, thus reducing data dimensions limitations.
For businesses, ModelOps is the key capability for scaling and governing AI and machine learning at the enterprise level. It serves as a collection of tools, technologies, and best practices for deploying, monitoring, and managing machine learning models effectively.