Using AI as a practical layer across research, design systems, prototyping, documentation, and implementation support. The goal is not to replace design judgment, but to make product teams faster, more structured, and more aligned from idea to shipped experience.
Figma AII use Figma AI and Figma Make as a design-system accelerator: exploring flows, generating structured screens, testing content states, and guiding outputs with libraries, tokens, variables, and reusable components so AI stays close to the product language.
CodexI use Codex to bridge product design and engineering: translating interface decisions into implementation tasks, reviewing front-end behavior, improving responsiveness, validating UI details, and keeping design intent visible through the build process.
ClaudeI use Claude for long-form product thinking: comparing journeys, structuring requirements, analyzing edge cases, reviewing complex flows, and turning scattered notes into clear decision documents for product, design, and engineering teams.
I use AI to synthesize interviews, analytics notes, stakeholder input, support themes, competitor patterns, and market signals into sharper opportunity areas. Then I turn those outputs into hypotheses, product principles, journey risks, and questions the team can validate.
I connect AI work to design foundations: components, variables, tokens, spacing rules, accessibility standards, UX writing patterns, and state coverage. This helps teams generate and evaluate options that respect the system instead of creating disconnected one-off screens.
I use Figma Make, AI-assisted prompting, and rapid iteration to move from concept to interactive prototype quickly. This makes it easier to compare flows, test alternative information architectures, identify missing states, and align stakeholders before expensive implementation starts.
I use Codex and Claude Code-style workflows to inspect interfaces, clarify behavior, prepare implementation notes, check responsiveness, document component logic, and review the user experience after it reaches code. This reduces the gap between design handoff and real product quality.
AI helps transform messy input into clear options, risks, and decisions, so teams spend less time waiting for first drafts and more time improving direction.
AI becomes more useful when it is connected to tokens, components, variables, content rules, and interaction standards rather than used as a generic idea machine.
I use AI to check edge cases, empty states, error states, accessibility considerations, UX copy, and responsive behavior before issues reach production.
AI supports better specs, clearer implementation notes, front-end review, and faster iteration between design intent and production reality.
I use AI to widen exploration, reduce repetitive analysis, improve documentation, and speed up delivery conversations. Final product decisions remain grounded in user evidence, business priorities, technical constraints, accessibility, brand standards, and design craft.
I treat AI prompts, examples, design-system rules, and review checklists as operating assets. They help teams reuse good judgment, keep outputs consistent, and make AI collaboration repeatable across product squads.