Modeling data in complex domains -- financial services, enterprise systems -- is a harder problem than even the people doing it often realize.
I write about what I think is an underappreciated part of that problem: building systems and data models that fully represent the domain and how it actually works. In my experience, teams working from semantically inadequate models often think they have data quality issues, when the real problem is that they haven’t modeled their business quite right.
AI is going to accelerate all of this, for better and worse. I think a lot about where it will help us build better models, and where it will just scale the problems we already have.
Some of what I write will be analysis. Some of it will be half-formed. I’m hoping some of it will be worth your time.
