We dive into why most AI tools today feel like “horseless carriages”: retrofitted instead of reimagined.

In this episode, we explore:

* The shift toward agentic software

* The need for headless systems and declarative interfaces

* Why precision and recall matter more than ever in data analytics

* Why evaluating agentic systems requires new abstractions drawing on physics, semantics, and statistical observability

* How AI operates in a kind of semantic space.

00:00:00 Introduction to Agentic Data and Skeuomorphic AI

00:04:26 The Shift to Headless and Agentic Software

00:14:18 Precision, Recall, and the Taxonomy of Questions

00:22:58 Challenges in LLM Evaluation and Industry Mindset

00:29:29 Physics Analogies for Agentic Software

00:38:28 Conclusion: The Future of Agentic Analytics

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit agenticdata.substack.com