Trust issue
Need AI, but team won't trust black boxes
Trust requires visible sources and refusal behavior
Full-Stack AI Engineer
I spent 18 years in advertising.
Then I learned to code.
Now I build AI products while keeping my day job as VP—and I write about the journey. The wins, the failures, and everything in between.
4+
Years Learning
512
Blog Posts
4
Live Products
From brief to system
The messy part is the proof.
Advertising taught me to read ambiguity: audience tension, client pressure, channel constraints, and the gap between an idea and something that ships. Now I turn that judgment into AI products and practical systems.
System gap
Too many channels, no shared operating model
Media work needs an operating model, not a prompt list
Keep judgment
Clients want speed without losing judgment
Human judgment stays in the loop by design
Review points
Content quality drops when workflow gets automated
Quality improves when workflow has review points
Proof gap
Hard to prove what changed after training
Progress needs observable milestones
Practice wins
Operators need practice, not another AI keynote
Training becomes a course plus practice system
Sydney answer card3 sources
Grounded answer
Sources visible before the recommendation.
Source-backed responseCourse access stateLesson 4
Operating model
Module progress turns behavior into practice.
Prova sprintDay 12
Sprint board
One active task, one reviewed task.
Ad operator
Strategy, media systems, and client reality.AI builder
Products that turn judgment into repeatable workflows.Operator systems
Patterns teams can apply as they build.What I've Built
Start with the course for the operating model. Use Prova to practice it through guided sprints. Use DIALOGUE to turn expertise into multilingual podcasts, and STRATUM to pressure-test strategy with AI frameworks.
Explore the Course
$199AI-Native Media Operations: From Workflow to Operating Model
What to automate. What to protect. Where to go deeper. How your team needs to change.
7 Modules~3 Hours16 Templates
•
The 75-80% AI-first operating model for media teams
•
Real failure cases, honest ROI, ethics guardrails
•
Module 1 is free — no purchase required
Explore the Course
$199
From Workflow to Operating Model
Start here
Pick the next step that matches how you want to learn.
Course if you want the full system. Paths if you want a curated starting point. Sydney if you already know the question.
Latest Posts
Thoughts on AI, learning, and building
•AI
I Rolled Back 2,724 Lines After One AI Audio Change Broke Production
I pushed what should have been an upgrade to my podcast platform's voice system. Six days and several commits later, I deleted 2,724 lines of code and rolled back to what worked. Here is what happened and what it taught me about testing production AI changes.
·AI
I Rolled Back 2,724 Lines After One AI Audio Change Broke Production
I pushed what should have been an upgrade to my podcast platform's voice system. Six days and several commits later, I deleted 2,724 lines of code and rolled back to what worked. Here is what happened and what it taught me about testing production AI changes.
·AI
I'm Using Claude Code for Everything Else But Coding
Nineteen days after cancelling Claude Max, my pattern has settled. Codex with GPT-5.4 on xHigh took the coding seat. Claude Code with Opus 4.7 on xHigh took every other seat at the desk.
·AI
Two years after my 7 Andrew Ng courses: the 2026 path I'd actually take
People still message me asking if the 7 Andrew Ng courses I recommended in 2023 are the right path today. Short answer: most of them, yes, but with a different roadmap around them. Here is my 2026 update, with per-course verdicts and a forked path for builders and operators.