Nine AI-era skills arranged by cognitive fit — not alphabetically, not by market demand, but by how naturally each one maps to how I think. The top five are the core zone: where intuitive, pattern-first, holistic cognition has a structural advantage.
★ Prime Skill
2
Skill 01
Evaluation & Quality Judgment
"Does this actually work — or just look like it does?"
Natural Fit
Pattern-first intuition: feels wrong before you can say why
This is the cognitive act I've been doing my whole life — in paint, in design, in client work. I sense when something is confidently wrong. Most people are fooled by fluency. I'm not. That immunity to fluency-as-competence is the rarest thing in AI right now.
Roles this unlocks
AI Output Evaluator
RLHF Trainer
AI Quality Analyst
AI Red Teamer
LLM Evaluator
Abstract painting = trained eye for wrongness
A great eval is one where multiple people look at it and reach the same pass/fail conclusion. I've been building that consensus sense — through critique, through client work, through every time I looked at a canvas and knew something was off before moving a single mark.
H
Skill 02
Human-AI Interaction Design
"Design the felt experience of working with AI."
Natural Fit
The prompts, feedback loops, moments of handoff, escalation to humans — this is UX work. The medium changed, the discipline didn't. Where my design background meets the new medium.
Roles this unlocks
AI UX Designer
Conversational UX Designer
Human-AI Interaction Designer
X
Skill 03
XR & Spatial Design
"Designing the impossible interface."
Natural Fit
Building hand-based AR UIs on Magic Leap 2 — where the interface is your hands, the canvas is the air, and every gesture is a spell. Geometric abstraction as UI. Abstract painting as spatial composition. This is where art and engineering converge.
Roles this unlocks
XR Experience Designer
Spatial UX Designer
AR/MR Interaction Designer
4
Skill 04
Context Architecture
"Build the library agents can actually search."
Natural Fit
Composition at a systemic level — what's foregrounded, what's retrieved on demand. I already do this in every layout. The NOOSPHERE project is context architecture made spatial and walkable.
Roles this unlocks
Context Architect
AI Information Designer
Knowledge Systems Designer
P
Skill 05
Product Thinking
"Does this actually work for a real person?"
Natural Fit
Every role I've held required asking whether the thing I was building worked for a real person. That question doesn't change in AI. It becomes more important.
Roles this unlocks
AI Product Designer
AI Experience Strategist
AI Creative Director
Growing & Stretch Skills
Active development areas — honest about the gap, grounded in a clear path
6
Skill 06
Trust & Security Design
"What's the worst that could happen — and to whom?"
Growing
I've designed for humans who make mistakes, who are confused, who are vulnerable. That caring instinct is the seed of trust design.
Roles
Responsible AI Designer
AI Accessibility Lead
7
Skill 07
Specification Precision
"Say exactly what you mean. Nothing left to inference."
Growing
A stretch for intuitive thinkers — but I already know what it feels like when a client misunderstands a vision. That knowledge of the gap is the foundation.
Roles
Prompt Designer
AI Experience Designer
8
Skill 08
Multi-Agent Decomposition
"Break work into what agents can actually hold."
Growing
I figure out project structure by feel — sensing where one thing ends and another begins. Making that natural sense of rhythm explicit for machines.
Roles
AI Product Lead
AI Workflow Designer
9
Skill 09
Cost & Token Economics
"Is this worth building an agent for?"
Stretch
The furthest from my natural style — but not a blocker. Build a spreadsheet calculator, plug in variables. I've estimated project costs without formal training. Same principle.
Senior path
Senior AI Product Manager
AI Architect
