Jul 10, 2026 · 31 min · 8 segments
In this episode, Ole Olesen-Bagneux sits down with Lulit Tesfaye, Joseph Hilger, and Zach Wahl from Enterprise Knowledge to explore why the future of enterprise AI depends on connecting knowledge…
Lulit TesfayeGuest
Ole Olesen-BagneuxHost
Zach WahlGuest
Joe HilgerGuest
Emma McGrattanHostAnd to a certain extent, some of my questions have already been answered, but I have many more.
But first of all, I think I would like to clarify for the listeners that I find that you guys in enterprise knowledge, you have a really interesting way of defining a semantic layer.
Because typically, technology vendors and the standard reference books on textures, they will describe diagrams.
And in the diagram, somewhere in the top will sit a box that says semantic layer.
But there is something very different about the reality of semantics in enterprise.
And I think you capture that very nicely, both in your book and the way you work.
So I don't know who wants to answer this one, but I would like you to unfold how you define a semantic layer and why.
I would say, and I alluded to this earlier, I think there is, if you look up a semantic layer, it's likely that you're going to hit the find a lot of definitions for it out there in the world.
And the biggest thing I would call out is semantics or the discipline of the semantic web is not new.
And the key thing here is when we talk about the semantic layer, it is not about the metric layer or the physical layer.
It's about the aboutness of your collectively knowledge assets, structured data, unstructured data, expertise, processes, anything you can attach a meta data to.
So by our definition and largely, I think, by the standard definition that's based on semantic web standards, semantic layer is about these aboutness components, business glossary that provides information definitions for your organization.
Revenue for a coffee company is not the same revenue as a bank or a regulator.
There is taxonomies that allow you to create these controlled vocabularies for what goes under our services.
Ontologies take that a little bit further and make the relationship between how a product is related to a customer and related to a specific service that you may provide.
And to a certain extent, some of my questions have already been answered, but I have many more.
But first of all, I think I would like to clarify for the listeners that I find that you guys in enterprise knowledge, you have a really interesting way of defining a semantic layer.
Because typically, technology vendors and the standard reference books on textures, they will describe diagrams.
And in the diagram, somewhere in the top will sit a box that says semantic layer.
But there is something very different about the reality of semantics in enterprise.
And I think you capture that very nicely, both in your book and the way you work.
So I don't know who wants to answer this one, but I would like you to unfold how you define a semantic layer and why.
I would say, and I alluded to this earlier, I think there is, if you look up a semantic layer, it's likely that you're going to hit the find a lot of definitions for it out there in the world.
And the biggest thing I would call out is semantics or the discipline of the semantic web is not new.
And the key thing here is when we talk about the semantic layer, it is not about the metric layer or the physical layer.
It's about the aboutness of your collectively knowledge assets, structured data, unstructured data, expertise, processes, anything you can attach a meta data to.
So by our definition and largely, I think, by the standard definition that's based on semantic web standards, semantic layer is about these aboutness components, business glossary that provides information definitions for your organization.
Revenue for a coffee company is not the same revenue as a bank or a regulator.
There is taxonomies that allow you to create these controlled vocabularies for what goes under our services.
Ontologies take that a little bit further and make the relationship between how a product is related to a customer and related to a specific service that you may provide.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
