Microsoft Intune design system
When I joined Microsoft Intune, the product was in the middle of a major visual and technical transformation. My role was to own and maintain Intune’s local design libraries and help the team navigate a transition across three different design systems, without slowing down the designers and engineers building the product.
The goal wasn't simply to adopt a new library. It was to make the transition feel seamless.
BACKGROUND
Intune started on Ibiza, a framework built for Azure products and optimized for the scale and performance requirements of Intune. As the product modernized, it began transitioning to Fluent UI. This introduced new opportunities for design tokens, greater consistency, a more scalable component foundation and better alignment between design and engineering.
But the story didn't end there. A few months later, Microsoft Security introduced SFE (Security Fluent Experience) as the design language that would eventually unify the broader security portfolio. That meant Intune had to migrate again.
GOALS
Align with engenieering on available components in production and those coming down the pipeline.
Provide enough documentation and guidance for teams.
Build templates, not just components to help designers move faster.
THE PROCESS
02
Make the system understandable
We organized the libraries and created guidance around which components applied to Intune, where Intune-specific patterns were needed, accessibility requirements and specifications, component usage and behavior.
03
Design with engineering, not for engineering
We partnered closely with engineering to keep our design libraries aligned with what was available in production and what was coming next. Together, we addressed gaps that required new development, and determined where Intune needed to deviate from the system.
04
Connect design systems to AI-assisted workflows
As AI-assisted design became part of our workflow, we extended the same foundation into our shared UX repository. We added templates and structured components that could be used with VS Code and Copilot, making it easier to translate established design patterns into code and prototype experiences faster.