Agents plus governed data

Analytics that answers the follow-up question.

An AI agent can do an analyst's work: read a question, find the data, run the query, check the result, and try again. What decides whether it is right is not the model. It is whether the data underneath has one definition, one copy, and one place access is enforced.

Either half alone gets you a demo.

Agents without governed data are fluent and wrong, at machine speed, with nobody checking. Governed data without agents is a well-run warehouse that still answers questions in a week.

Put them together and the long tail of questions, the ones that were never worth an analyst's time, becomes answerable. That is the whole proposition, and most of the work is on the data side.

Start here

What the term means, how it differs from the dashboard era, and why the earlier attempt at this stalled.

The other half

Why an open lakehouse

Agents multiply readers, multiply questions, and multiply the temptation to make a private copy. An architecture where the data sits in open formats, with one catalog deciding access, gets better under that pressure. One where reading requires a particular vendor's engine gets more expensive.

Five Apache projects settle five separate questions. Each one is a specification with more than one implementation, which is the property that makes the stack assemblable at all.

All seven substrate entries

Running it for real

The part that decides whether a pilot becomes a capability: identity, evaluation, cost, and the order to do the work in.

Longer reads

Published pieces that carry parts of this argument in more depth.

Everything else, including video and community

Dremio blog The working archive: agentic analytics, semantic layers, catalogs, and the lakehouse underneath all of it. Read the archive YouTube, data and AI Walkthroughs and explainers covering both halves of this subject, from agent tooling to table format internals. Watch Data Lakehouse Hub The community hub: articles, a knowledge base, events, and a Slack where practitioners compare notes on what actually shipped. Join the community

Two newsletters, one list

Keep up with both halves of this.

Agentic analytics moves on two fronts at once. One newsletter covers the model and agent side, the other covers what the Apache projects underneath are shipping. One subscription gets you both, and neither costs anything.

Subscribe on Substack