AI product strategy
Choose a workflow worth solving.
Clarify the user decision, source data and acceptable failure modes. Define which work belongs to a model, a deterministic system or a human reviewer.
AI advisory and technical strategy
Turn AI ambition into
engineering decisions.
I bring the perspective of an engineering leader, independent AI builder and technical author. I help frame the architecture, evaluation and operating choices that determine whether an AI product deserves its users’ trust.
- 50
- people led through
three engineering managers - 2,000+
- engineers enabled
across 100+ teams - 100+
- independent apps
across platforms - 20+
- AI books and
engineering field guides
Where I can contribute
Make the next decision
with a clearer technical picture.
My advisory interests center on AI products, developer tools, workflow automation, consumer commerce and privacy-aware software.
Agent architecture and evaluation
Make autonomy measurable.
Review tool permissions, context and memory, orchestration, evaluation scenarios, observability and recovery. Define evidence for a release decision.
Engineering adoption
Turn a tool into a team capability.
Connect AI-assisted development to code review, automated checks, engineering standards and manager accountability. Draw on platform adoption and organizational change experience.
Privacy and governance
Set boundaries before deployment.
Identify sensitive data, model access boundaries, approval points and audit needs. Examine whether the architecture supports the product’s promises.
Why this perspective
Experience on both sides
of the architecture decision.
At PayPal, I lead 50 people through three engineering managers and own production Shopping AI/ML integration and operations. Independently, I build the controls, evaluations and product workflows I write about.
Organizational scale
PayPal UI adopted by 2,000+ engineers across 100+ teams, alongside shared infrastructure handling 50M+ daily requests.
Implementation depth
KaiInbox source validation, Sanctuary privacy controls, agent-workflow experiments and educational evaluation tooling.
Published engineering
20+ AI books and field guides within a broader 25+ book catalog, covering agents, context, AI-native delivery and governance.
Start with the problem
Bring the decision
you are working through.
I welcome selective advisory, technical diligence and engineering-leadership conversations. Share the product context, the engineering decision and the outcome you want to improve.