Assist and AI Insights closing the feedback loop

To streamline customer feedback for home improvement professionals, I designed an end-to-end AI ecosystem that bridges the gap between data insights and proactive resolution. By leading a WCAG-compliant redesign of our AI Insights dashboard, I moved beyond color-reliance to use universal iconography and standardized color tokens, ensuring all users can identify at-risk customers at a glance. This logic feeds directly into ✨Assist, a responsive AI-powered toolbar I developed that features shortcut-based editing and a "subtle-first" UI placement to minimize friction. Since its launch to 6,000+ clients, this system has reduced average response times by 80% and established a scalable, reusable design pattern for AI generation across the entire organization.

Accomplishments

  • Inclusive Design Systems: Redesigned global color tokens and iconography to meet WCAG AA standards, ensuring data accessibility for color-blind users while improving overall legibility.

  • AI-Driven Interaction Design: Leveraged competitive benchmarking to design a non-intrusive "subtle-first" AI toolbar, integrating shortcut-based editing patterns to solve the "blank page" problem for users.

  • Scalable Architecture & Prototyping: Developed a fully responsive, multi-functional component pattern using Claude Code to audit and manipulate stylesheets, saving months of future development across the product suite.