Design technologist working on AI evaluation and accessibility. Safe AI requires accessible AI.
I test whether AI systems do what people need, including the people a test usually leaves out, and I build the open-source tools that make that measurable.
Projects
- klar: color accessibility CLI for people and AI agents. Includes a pre-registered eval of how agents receive written guidance.
- OKCA: open-source contrast algorithm that never approves a color pair WCAG 2 rejects.
- Candor: design system with contrast minimums in its tokens, checked on every pull request.
Writing: pawn002.github.io/blog-public. Start with Written but Unread: Do AI Agents Read CLAUDE.md?



