Efficient MLOps: Developing and Deploying ML Models with Databricks

Explore the design and implementation of an efficient MLOps framework for successful ML model development and deployment in this 28-minute conference talk. Dive into managing diverse environments, implementing robust CI/CD pipelines, and adopting standardized coding practices. Learn about tools for ensuring code quality and test coverage. Discover the strategic use of Unity Catalog, MLflow, and MLOps stacks for managing and monitoring model evolution across various stages. Examine advanced techniques like Delta Live Tables and Lakehouse Monitoring for robust input and output data drift monitoring to maintain model consistency. Gain insights into creating a standardized approach for ML model development across multiple countries. Watch a demonstration showcasing the framework's potential, presented by Alessandro Mazzullo and Lavinia Guadagnolo from Plenitude.

Efficient MLOps: Developing and Deploying ML Models with Databricks