AI LAB: Experimentation = Relevancy
I created this experimentation lab to explore UX, AI, and product design in a way I couldn’t find elsewhere — practical, hands-on, and transparent. I learn by reading and doing. Along the way, I’ll share short, concise case studies documenting what I build and what I learn.
Core tenets
These tenets help keep me focused, experimental, and accountable — structured enough to create momentum, flexible enough to allow discovery.
Use AI Intentionally
Use AI tools across the full spectrum of exploration — notation, ideation, research, design, prototyping, and the unknown.
Experimentation Over Perfection
If I end up in a cul-de-sac, that’s okay. I’d rather have taken the trip then not attempted the journey at all. Every detour teaches me something.
Use Domain Knowledge
Focus on travel. I know the ecosystem, the users, the edge cases, and the competitive landscape. Familiar context allows me to experiment faster and push ideas further.
Don’t Boil The Ocean
Play with common travel problems that need solving, but don’t get bogged down unnecessarily. Play, learn, share. Repeat.
Make It Inclusive
Inclusive design is built in, not layered on. Experiments should account for assistive technologies and real-world constraints.
Reflect Before Resetting
At the end of each experiment, I step back. I compare where I started to where I finished, even if it’s incomplete.
The findings
The deeper I get into the AI ecosystem, the more the Lab becomes. A published point of view on what AI is doing to the creative industry. A vibe-coded app - Trip Atlas. Another app on its way. List order most recent first.
#10 Gen AI POV
A point of view on Generative AI and the creative industry — built from eight structured experiments, a live product build, and the observed creative reckoning happening across every industry simultaneously. A designer's field notes from inside the disruption.
Read the POV here
Trip Atlas is a vibe coded web-app I designed and built as a travel photography planning tool for a 3-week, 4-city trip to Edinburgh, Amsterdam, Barcelona and Dublin this year. Built with Claude, v0 and Github. Updates and optimizations are currently planned.
Read the Trip Atlas Case Study
Explore Trip Atlas (the app)
My first major concept ideation with Claude. In this experiment I outline an idea for a traveler concierge app to be outputted as a comprehensive plan for implementation. Along the way the idea evolved to include additional scenarios and specifications, but concluded under the target session rate.
AI app: Claude AI
#7: DS Architect
I asked Claude Sonnet 4.6 to act as a Design System Architect and prompted it to create a travel company design system (referencing Expedia) outputting design principles, foundations, components, patterns, JSON file, and developer guidelines.
View the published document
From structural reconstruction and dynamic module generation to accessibility enforcement and semantic annotation, this experiment evaluates not just output quality — but judgment.
AI app: Loveable
Theorizes that different AI design agents can generate the same or nearly similar experience starting with the same screen artifact and prompts.
AI app: Figma Make
Building on Experiment 3, I prompted the agent to generate accessibility annotations using a prescribed structure and methodology to evaluate efficiency and accuracy.
AI app: v0 by Vercel
This experiment marked my first deliberate accessibility review using assistive technology on a live interface. As someone primarily focused on front-end design, this was an opportunity to move beyond structural assumptions and validate decisions through real interaction.
AI app: v0 by Vercel
Recreates a system-based lodging page in live code using v0, then introduced dynamic traveler-specific modules to test layout mechanics and interaction behavior.
AI app: v0 by Vercel
Tests whether a static system screenshot could become a functional proof of concept using AI. The experiment shifts from replication to controlled, audience-driven iteration.
AI app: UXPilot
About AI Lab
The goal is consistency over intensity. Each experiment isolates a specific design question and tests how AI tools can accelerate,challenge, or reshape the workflow. I’ll revisit similar problems across different AI platforms to compare outputs, assumptions, and integration patterns.
My goal isn’t just to understand these tools, but to use them in practical scenarios so I can speak about them with clarity and confidence. Hands-on application helps me see where they’re effective, where they fall short, and how they fit into real workflows. It also keeps me current as the tools and expectations continue to evolve.