This is the first post on the Rational Data Services blog. Going forward, this is where we'll share short, practical notes from the field — the kind of thing we learn building governed data products for federal and regulated clients.
What to expect
A few themes we'll keep coming back to:
- Data Fabric vs. lakehouse — when a deterministic, federated query layer beats centralizing everything.
- Governance that holds up — Need-to-Know enforcement with OpenMetadata, Ranger, OPA, and Trino.
- Agentic AI on real data — how to put a trustworthy, auditable data layer beneath an MCP interface.
Good data products make the right answer the easy answer — for people and for agents.
If any of this is relevant to what your team is working on, get in touch.