The Superintelligence Podcast

How AI is changing product design

Alexander Danilowicz

Alexander Danilowicz

CEO, Magic Patterns

Magic Patterns

“Lots of AI tools look great in demos but aren’t helpful when you try to ship real features.”

In this conversation

How does an AI design startup go from raising around $100,000 at Y Combinator Demo Day to a $6 million Series A? Alexander Danilowicz, CEO and co-founder of Magic Patterns, joins Superintelligence Editor-in-Chief Kim Isenberg to discuss the company’s text-prompt-to-web-app tool for product teams. Danilowicz describes the pivots that came first, from a design token manager to a component library editor called Mirrorful that customers paid for but barely used, and an MVP built on GPT-3.5 in about a week. He explains why Magic Patterns stays front-end only and builds on a team’s own components and design system, imported from Storybook or Figma, which he says sets it apart from v0, Lovable and Bolt. He also recalls how product managers surprised the team by valuing the visual mockups more than the code. The conversation also covers which models the product runs on, the margins of AI tools, and why he expects user interfaces to remain even with AGI.

Key takeaways

  • Magic Patterns raised a $6 million Series A from Standard Capital and is hiring quickly.

  • The company crossed $1 million in ARR over the summer, after a year of constant pivoting at Y Combinator with zero revenue, Danilowicz says.

  • Compared with V0, Lovable or Bolt, Magic Patterns focuses on integrating a team’s existing component library and product styles.

  • Sonnet models are not cheap, Danilowicz says. The company was profitable for a while and is not while scaling, but its margins are still good.

  • The biggest surprise was how strongly product managers gravitated toward the visual mockups rather than the code.

The daily AI briefing people actually read.

What shipped, who’s winning, and what it means. Six days a week, in under ten minutes.

The newsletter

The people building the AI future. Read by 160,000+ subscribers.

Subscribe free