Notes on agentic data work.
Field notes from the founders on industrializing the last mile of data — governed, tested data products, automated governance, and the shift to AI-native platforms.
-
From raw to golden data: the journey your data actually takes
Ingestion is the easy part. Between the moment raw data lands in your lakehouse and the moment anyone can trust it, there is a long, manual journey — profiling, cleansing, modeling, mastering. This is what that road looks like, and what happens when agents walk it for you.
Read -
White paper — NDMO compliance by architecture: mapping the 15 domains to an agentic data platform
Saudi Arabia's NDMO standards span 15 domains, 77 controls, and 191 specifications. This paper maps every domain to platform architecture — what can be automated, what can be accelerated, and what stays organizational — and argues that the difference between a compliance program that survives and one that dies is architectural.
Read -
We industrialized everything around the data. Not the data.
In twenty years we made data almost free to produce, cheap to store, and fast to move. The work of turning it into something you can trust still happens the same way it did in 1995 — by hand.
Read -
The missing layer: a practitioner's case for an AI Data Fabric
Two decades of data platforms taught me the problem was never the tools. It's the gaps between them. Here's the case for an intelligent layer that connects what you already have, with governance built into the architecture instead of bolted on after.
Read
Prefer a feed? Subscribe via RSS.