The Lost Art of Data Modeling (And Why It’s Making a Comeback)

What do knowledge graphs, beautiful wine, and the AI revolution have in common? According to Juan Sequeda, Principal Scientist at data.world and co-host of the popular Catalog & Cocktails podcast — everything.

In this episode of How I Met Your Data, Juan joins Angela and Karen for an energizing, thought-provoking deep dive into why semantic data modeling is having its renaissance moment, what we’ve misunderstood about knowledge-first design, and how data governance might just be on the brink of being...cool??

From dishing on the history of data modeling’s pendulum swings to exploring how ontologies and entity-relationship diagrams can fuel AI effectiveness, Juan breaks it down with clarity, humor, and plenty of passion (plus a little wine talk for good measure 🍷).

Tune in to hear:

• Why “data modeling” isn’t dead—it just went underground

• How semantic modeling helps LLMs deliver better, more accurate results

• The importance of pushing semantics left to reduce tech debt

• What makes a data product actually reusable (hint: not just pipelines)

• Why governance might finally be cool—and how to make it a true enabler, not a bottleneck

Whether you're semantic-curious, governance-skeptical, or just trying to make sense of your enterprise data chaos, this is the episode that’ll have you rethinking your metadata strategy—and maybe even opening a bottle while you're at it.

This one’s for the data dreamers, the AI realists, and everyone stuck in the modeling middle.