Four distinct systems built over six years. Each one solved a different scale problem, but the question behind all of them was the same: what does the system need to make quality hold when no one is watching? What designers build in Figma, what AI generates in production, what customers see in their published sites, each system was built to close that gap without requiring review cycles or proximity to any single person.
01
Hub design language
Tokens, type scale, and components built for the professional segment. Durable enough to survive a 360K customer migration intact.
2020
02
Mobile native system
Foundational strategy for GoDaddy's cross-product mobile ecosystem. Gained VP visibility and became the basis for 2026 company-wide initiatives.
2024
03
Site output quality
Design assets and standards for improving the quality of websites the builder products generate. Directly tied to AI builder adoption metrics.
2025
04
Agentic builders system
Shared language model, component library, and AI output quality standards spanning three builders and three customer audiences.
2025
From Guidelines to Embedded ConstraintsFrom Implicit Knowledge to Explicit System Context
The purpose of a system is not consistency. It is to make quality independent of who is in the room. If it only holds when someone is watching, it is not built to last.
Hub design language
Common language model
Mobile native
Site output quality
Site output quality
Product impact
Antares WSB migration — resolved seven years of design system tech debt, creating the shared infrastructure that made the agentic builders system possible.
Piloted dark mode as the first GoDaddy product to use it.
Common Language model — aligned research, content, and design around real customer vocabulary; reduced AI drift across three builders simultaneously
Site output quality standards raised the baseline of what AI generated, part of the system work that lifted AI feature adoption 99% across builder products.
Mobile strategy culminated in company-wide 2026 initiatives after gaining VP-level visibility and cross-org reach
Each system embedded constraints upstream so quality held without review cycles or proximity to any single person
Team and org impact
Mobile work elevated a direct report's scope from team-level to org-wide initiative — I created conditions for the work to hold without me as owner
MCP-driven standards replaced review cycles as the quality gate — quality became a system property, not a person property
Systems practice became the model for how GoDaddy UX operates at scale — other teams built on and extended the infrastructure
Each system built to be teachable and delegatable — a new team member could pick up constraints without re-learning the reasoning behind them
Common Language unified research, content, and design as one discipline for the first time, reducing siloed decisions about AI behavior
"
EJ fosters a culture of trust, allowing his team to take ownership while providing support when needed.