Getting a model into the real world and generating business value from it involves more than just building it. How to deploy, serve, and inference models at scale? How to continuously monitor and deliver when you have
code changes, data drifts, or the model needs to be retrained? Automation and management of the end-to-end ML life cycles essentially is MLops. This talk will introduce DataOps tools like Delta Lake and Feature Store that track and version data changes, and mitigate online/offline inference skew. We will also use code examples to demonstrate how to deploy an image classifier with MLflow.

About the speaker:
Yinxi Zhangis a Sr. Data Scientist at Databricks with 7+ years of industry experience on end-to-end ML development. Her responsibilities as a Brickster are teaching Scalable Machine Learning, Deep Learning, and MLops courses and helping clients develop their ML solutions. She is also an Austin Deep Learning Meetup member. She used to be a marathon runner and now a yogi