Human Data Expertise
for the AI Era

Hands-on write-ups for architects and data engineers putting AI into production, especially in regulated industries. Each post reports a working build: what was built, what was measured, and what the test does not show.

20+years in IT and data
45posts from working builds
2clouds tested: Google Cloud and AWS
3running series

Latest posts

All posts →

Choosing an AI gateway operating model

The AI gateway decision is about how many data planes sit under one control plane, and model traffic and agent tool traffic answer it differently. This sets out the three operating models, a routing rule for placing traffic, and the standard a local gateway must meet to carry it.

Read →

Which parts of your eval framework you can stop maintaining

A managed evaluation service can take over the judge, but not the contract tests, the drift canary, or anything needing provenance for the judge itself. Testing one produced four faults that each returned a plausible number instead of an error.

Read →

Brain Breaks

Small browser games for when you need a minute away from the screen you were already staring at.

Get the next post by email

One email when a new post goes up. No tracking, unsubscribe in one click. Subscribers also get the RAG security checklist.

Need a second pair of eyes?

Independent reviews of AI gateways, RAG security, data governance, and Google Cloud data platforms, from the same hands-on approach as the posts.

See consulting services