How the declarative component model and AI-augmented development let me ship across web, desktop, mobile, email, video, and agent interfaces in a single week.

A clear breakdown of AI models, products, and agents — the three layers of the AI coding landscape and how they relate to each other.

Teams that succeed with AI are not just using better tools. They are working differently. Here's the workflow shift that matters.

After watching teams adopt AI tools for months, here's the sustainable workflow that sticks: plan mode + iteration + continuous verification.

When engineers complain AI tools are slow, it is usually not the AI. It is how they use it. Here are strategies for keeping Claude Code fast and responsive.

A breakdown of approaches to migrate from legacy systems to design systems, with insights from industry experts.

My take on deciding whether a component belongs in a design system. A practical guide for designers and product managers that looks at reusability, frequency, brand consistency, and maintenance.

My take on how flexible a design system should be, and the balance between local needs and global consistency.

Why I think semantic versioning does not fit Figma. Learn better ways to track design system changes.

Why boolean props feel natural in design tools but cause problems in React.

Flexible strategies for building polymorphic components in our libraries

Why composition leads to cleaner and more flexible components.