At the core, Machine Learning Operations (MLOps) turns an experimental machine learning model into a production system. MLOps is an emerging practice distinct from traditional DevOps. The ML lifecycle aggregates training data, making MLOps workflow sensitive to data changes, volumes, and quality. Additionally, matured MLOps should support monitoring both ML lifecycle activities and production model monitoring. Key Takeaways: • Build cloud-native serverless analytics pipeline. • Learning different data automation pattern including data lake, lake house and data mesh architecture. • Automate end-to-end data pipeline with MLOps.