That's an interesting comparison. Ignoring the LLM for a minute, I wonder if there are ways to build the tube through London dbt project. Like, if you end up with that kind organically created structure (which is really hard to avoid), how much can you put another layer on top of it?
The semantic layer needs a semantic layer, I guess.
That's an interesting comparison. Ignoring the LLM for a minute, I wonder if there are ways to build the tube through London dbt project. Like, if you end up with that kind organically created structure (which is really hard to avoid), how much can you put another layer on top of it?
The semantic layer needs a semantic layer, I guess.
Yeah something like that. In London there is an explicit map, but in a tech stack no system ever specifies the relation between a CRM and a CDP, it is just implied based on references etc (Tube map construct feels a bit ropy at this point!)
That's an interesting comparison. Ignoring the LLM for a minute, I wonder if there are ways to build the tube through London dbt project. Like, if you end up with that kind organically created structure (which is really hard to avoid), how much can you put another layer on top of it?
The semantic layer needs a semantic layer, I guess.
Substack didn't notify me on the response :/
Yeah something like that. In London there is an explicit map, but in a tech stack no system ever specifies the relation between a CRM and a CDP, it is just implied based on references etc (Tube map construct feels a bit ropy at this point!)