Search for a command to run...

Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
In this fascinating episode, Ryo Lu, head of design at Cursor, explores his design philosophy rooted in the mantra "it's all the same thing." (07:46) He views the world as fundamentally modular, where simple rules and patterns endlessly recombine to create emergent complexity. For Ryo, design is consciously participating in this process: seeing through surface complexities to understand underlying structures and rearranging them into new forms. The conversation delves into how AI is transforming design from feeling like painting in tools like Figma to sculpting clay in Cursor, allowing designers to work directly with code—the true material of digital products. (33:43)
Ryo Lu is the head of design at Cursor, where he's focused on building the next generation of tools for making software. Prior to Cursor, he spent five years as a designer at Notion, working across numerous projects including Notion AI. He has also worked as a designer at Stripe and Asana, contributing to some of the most influential software tools of the last decade. Growing up between China and Montreal, Ryo brings a unique perspective to design that bridges technical depth with philosophical thinking about how tools can amplify human creativity.
Ryo explains that true simplicity requires wrestling with complexity first—like a swan that appears serene on the surface while paddling intensely beneath. (11:44) You can't design elegant systems by starting simple and staying simple. Instead, you must test your conceptual models against real-world complexity, identify weaknesses, and iteratively refine until you achieve both surface elegance and robust underlying systems. This means allowing for experimentation, controlled chaos, and multiple iterations rather than trying to perfect solutions in isolation.
The most profound shift in Ryo's design practice has been moving from creating static mocks to working directly with code—the actual material of digital products. (38:28) As he puts it, "our material as software makers is never the pixels. It is the code itself that renders the pixels." With AI tools like Cursor, designers can now prototype by sculpting clay rather than painting pictures, getting immediate feedback from the material itself. This allows for faster iteration, better understanding of constraints, and solutions that emerge from the interaction with real systems rather than theoretical designs.
Rather than forcing users to choose between simple and complex, rigid and flexible, Ryo advocates for designing systems that accommodate the full spectrum of user sophistication. (58:00) The goal is creating tools that work beautifully for beginners while scaling to power user needs—allowing people to start simple but access deeper complexity as they grow. This requires thinking systemically about zero state, one state, and n state experiences, ensuring they all feel cohesive while serving different levels of user expertise.
The best systems have slack—controlled chaos that provides stability while remaining loose enough to evolve. (59:59) This means occasionally allowing "ugly" features or divergent solutions to exist temporarily, observing how they're used, and then deciding whether to clean them up or let them point toward new directions. Rather than optimizing everything immediately, strategic slack allows for experimentation and organic evolution while maintaining core system integrity.
Ryo believes there's often an "ultimate solution" for given constraints, but you can't think your way to it—you must build your way to it. (21:44) Truth in design comes from constant interaction between conceptual understanding and real-world testing. You start with core building blocks that don't change much, but you discover their ideal configuration through repeated cycles of building, testing, gathering feedback, and refining. The goal is finding solutions that feel retroactively inevitable but can only be discovered through this iterative process.