MLOps on Azure Databricks with MLflow

Explore MLOps on Azure Databricks with MLflow in this 53-minute tech talk by Oliver Koernig, Solutions Architect at Databricks. Learn how to implement an integrated MLOps lifecycle using Databricks' managed MLflow and the Azure ecosystem for managing and deploying machine learning models. Dive into the MLflow Model Registry, a centralized model store with APIs and UI for collaborative model lifecycle management. Get a detailed preview of the MLflow Registry Webhooks feature for automated MLOps pipeline triggering. Follow along with the demonstration using the provided GitHub repository. Discover topics such as Azure ML, notebooks, webhooks, pipeline automation, model drift detection, and result reporting. Gain insights into Databricks' leadership position in Gartner's Magic Quadrant for Cloud Database Management Systems and Data Science and Machine Learning Platforms.

Introduction
MLflow
Azure ML
Notebook
Webhooks
Running the pipeline
Demo
Testing
Model Registry
Pipeline
Automation
Azure MLL
Detecting Model Drift
Sending Results