25 min
Oct 11, 2022
8:50 am

Accelerating ML at Uber with the Palette Feature Store

Amit will show how the Michelangelo ML Platform uses the Palette Feature Store to address inefficiencies of the model lifecycle.

About this session

At Uber, we've observed over the years that building feature pipelines and iterating over them resulted in one of the highest friction in productionizing ML models. Furthermore, discovering issues with the data pipelines during training further delayed the production process. Finally, once models were ready to be deployed, provisioning the online feature store ended up being a major bottleneck.

In this presentation, we will show the Michelangelo ML Platform makes use of the Palette Feature Store to address these inefficiencies across various stages of the model lifecycle, vastly cutting down the time-to-launch across ML teams.

Moderator

Session Speaker

Session Speaker

Session Speaker

Session Speaker

Session Speaker

Join our Slack channel to stay up to date on all the latest feature store news, including early notification when the conference details emerge.