Rethinking the Semantic Layer, Part II: The Builders Respond

What's left for the semantic layer to do when agents read schemas, inspect code, generate queries, and follow lineage on their own? Part II of our semantic layer conversation, where three builders shipping these products respond. In Part I, Jacopo argued that most of what a semantic layer provides for agents already lives in well-written pipeline code. Here, the builders push back. Joining the conversation: - Matthew Mullins, CTO, Coginiti - Simba Khadder, Head of AI Products, Redis - Paul Blankley, Founder, Zenlytic Moderated by Jacopo Tagliabue, CTO, Bauplan Chapters: 00:00 - Welcome 04:30 - What is a semantic layer? 9:00 - Do agents still need query compilation? 21:45 - Trust, validation, and the memory layer 43:45 - The cost question Watch Part I: https://youtu.be/LeeE-J4hEM8?si=_sqZpwmp4C2whcNk Try Bauplan Skills: https://github.com/BauplanLabs/bauplan-skills Helpful Links: 📚 Official Documentation: https://docs.bauplanlabs.com 💻 GitHub Repo: https://github.com/BauplanLabs/ 🌐 Bauplan Homepage: https://bauplanlabs.com 📄 Technical Blog: https://www.bauplanlabs.com/blog 📣 Follow Us: Twitter: https://x.com/Bauplan_labs LinkedIn: https://www.linkedin.com/company/99837917/ #DataEngineering #AIAgents #Lakehouse #Python #GitForData #DataPlatform #ClaudeAI