"We'll add governance later" is the most expensive sentence in AI adoption.
My take after another audit post-mortem.
Read on LinkedInInsights
The thinking behind shipping AI into regulated systems.
I write the deep dives here first — canonical, unhurried, ungated. Then I share them, and the best of what I'm reading, on LinkedIn, dev.to and X.
Start here
Deep dive8 min read
The AI Audit Trail That Survives an Inspection: capture, seal, replay
You can't audit a model that retrains — you audit the trail. What to capture, how to seal it tamper-evident, and how to replay one decision when the auditor asks.
By André Queiroz · dezotech
More deep dives
One new deep dive in the pipeline each month.
Deep dive
Shadow AI Is Already in Your Org: a 90-day containment plan
What to find, what to fence off, and what to formalize — before unsanctioned AI shows up as an audit finding.
8 min read
Deep diveAdopt AI Without Failing Your Next Audit: a 2026 playbook
The exact frameworks, guardrails and shadow-AI assessment I run on real engagements — for the people who'll be in the room when the auditor asks "who approved this?"
9 min read
Join the newsletter
Subscribe and the next deep dive lands in your inbox before it hits LinkedIn.
Not articles — just the threads, posts and notes worth your time, hand-picked from where I work in the open. Links go out to the original.
Strangler-fig, but for compliance: migrating a legacy core with no downtime window.
The reference architecture I keep linking people to.
Read on dev.toA short thread on what the EU AI Act Aug-2 deadline actually changes in production.
The thread that got the most questions.
Read the threadShip & Stay Compliant
One practical playbook every other week.
Automation, legacy modernization & safe AI adoption. Free: the AI-Adoption-Without-Audit-Failure checklist.