Feature Stores for MLOps with Mike del Balso - #420
Published:Oct 19, 2020 15:02
•1 min read
•Practical AI
Analysis
This article is a summary of a podcast episode from "Practical AI" featuring Mike del Balso, CEO of Tecton. The discussion centers around feature stores in the context of MLOps. The article highlights del Balso's experience building Uber's ML platform, Michelangelo, and his current work at Tecton. It covers the rationale behind focusing on feature stores, the challenges of operationalizing machine learning, and the capabilities mature platforms require. The conversation also touches on the differences between standalone components and feature stores, the use of existing databases, and the characteristics of a dynamic feature store. Finally, it explores Tecton's competitive advantages.
Key Takeaways
- •The podcast episode discusses the importance of feature stores in MLOps.
- •Mike del Balso, the CEO of Tecton, shares his insights on building and operationalizing machine learning platforms.
- •The conversation explores the differences between feature stores and other components, and the competitive landscape.
Reference
“In our conversation, Mike walks us through why he chose to focus on the feature store aspects of the machine learning platform...”