Machine Learning Operations (MLOps) is not a buzzword anymore. In the last few years, there has been a lot of booms in different MLOps tools and frameworks. Basically, it’s a paradigm that focuses on the automation and operationalization of AI development, including model packaging, monitoring, and deployment. While there have been advancements in MLOps tools and frameworks, there are still challenges and research gaps when it comes to deploying MLOps pipelines on edge devices. This paper provides an overview of the Edge MLOps area. Our aim is also to define the basic architecture and components needed for the industrial MLOps deployment. In this context, we also highlight some tools and frameworks. Moreover, we address some limitations and research gaps in the Industrial edge MLOps area.