Deploying Machine Learning (ML) models in Next Generation Network (NGN) presents significant challenges, particularly in managing the entire lifecycle of model creation, deployment, and operation. Traditional methods lack integrated pipelines that address the diverse and complex nature of the network, leading to inefficiencies and delays. The NEMI framework employs MLOps pipelines designed to streamline the ML lifecycle in NGN environments. The pipeline facilitates efficient data collection, model generation, and deployment, towards ensuring scalability, automation, and continuous integration. The approach in this article showcases the pipeline’s potential to enhance performance and reliability in NGN domains, emphasizing the critical role of advanced MLOps pipelines in optimizing ML operations within complex network environments. Using anomaly detection in a core network (Open5GCore) as a practical example, we demonstrate the pipeline’s features.