Feature Platforms for Data-Centric AI with Mike Del Balso - #577
Research#AI Infrastructure📝 Blog|Analyzed: Dec 29, 2025 07:42•
Published: Jun 6, 2022 19:28
•1 min read
•Practical AIAnalysis
This article summarizes a podcast episode from Practical AI featuring Mike Del Balso, CEO of Tecton. The discussion centers on feature platforms, previously known as feature stores, and their role in data-centric AI. The conversation covers the evolution of data infrastructure, the maturation of streaming data platforms, and the challenges of ML tooling, including the 'wide vs deep' paradox. The episode also explores the 'ML Flywheel' strategy and the construction of internal ML teams. The focus is on practical aspects of building and managing ML platforms.
Key Takeaways
- •Feature platforms are a key component of data-centric AI.
- •The evolution of data infrastructure and streaming platforms is crucial.
- •Building effective ML teams requires addressing specific organizational challenges.
Reference / Citation
View Original"We explore the current complexity of data infrastructure broadly and how that has changed over the last five years, as well as the maturation of streaming data platforms."