Rethinking the Semantic Layer in an Agent-Native Data Stack

The semantic layer existed because humans could not read SQL. Agents can. Jacopo Tagliabue (CTO, Bauplan) makes a contrarian case: most of what a semantic layer provides for agents already lives in well-written pipeline code- decorators, type hints, Pydantic classes, and docstrings, persisted next to data in Iceberg. Chapters: 00:00 - Introduction 07:27 - What the semantic layer is and why it exists 14:17 - The contrarian thesis: LLMs and open formats change the equation 17:21 - Example: revenue by country, in code 40:25 - Q&A Try Bauplan: https://app.bauplanlabs.com/ 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