Decoding Data Automation With Cloud-Native Analytics Pipeline and MLOps

Presented by

Saurabh Shrivastava, Global Solution Architect Leader, AWS

About this talk

At the core, Machine Learning Operations (MLOps) takes an experimental machine learning model into a production system. MLOps is an emerging practice distinct from traditional DevOps. ML lifecycle involves using patterns from 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.

Related topics:

More from this channel

Upcoming talks (3)
On-demand talks (615)
Subscribers (84084)
Data is the foundation of any organization and therefore, it is paramount that it is managed and maintained as a valuable resource. Subscribe to this channel to learn best practices and emerging trends in a variety of topics including data governance, analysis, quality management, warehousing, business intelligence, ERP, CRM, big data and more.