Ryo Lu
Related people
Episodes
Notes
First of Kind · Ryo Lu: The Way
-
Does AI Distribute or Concentrate Power?
Ryo Lu says AI both levels the playing field and raises the ceiling at the same time. Soleio had asked which way the power moves: an older hope that technology would push power out to the edges, or a new concentration where a 10x engineer becomes 1,000x more productive, often with Cursor. Ryo Lu does not pick one side.
Before these tools, doing anything meant learning the programming concepts, the syntax, and how to run the server. Now anyone with an idea can start. The first output can be a little sloppy, purple gradients, and more iteration brings it closer to what that person wanted. Telling an agent the exact thing still takes knowing the language, the concepts, and sometimes reading th
-
How Many AI Agents Can One Programmer Run at Once?
Ryo Lu says the best programmers he knows can maybe run only four agents at once. A person looks at one chat thread at a time and cannot take in that much information at once. The problem that follows is how to help someone who would otherwise be chatting with 15 agents at once or reviewing every change those agents make.
Cursor's interface started as manual coding, files and lines typed by hand. Completion came next: the product finished the next function, jumped to the right file, and tried to keep the person in flow while they typed. Chat made the model aware of the whole codebase and able to answer questions. Editing let it apply changes, and tools made the work more agentic. People are
-
What Is Design at Cursor?
Ryo Lu says design at Cursor right now is four people, including him: two people on the product and one person on the brand. Almost all the engineers are designing too. They start building, and they design the system as they build it, even when they care less about the visuals or the flow at first. His job is to wrangle that work back, clean it up, and unify the concepts so the product stays cohesive. The work is a lot more chaotic, and that makes it more fun.
The day is both thinking ahead and trying ideas. Thinking ahead asks where Cursor should get to, given what people do today and the systems already in place. New ideas can sit a little further out or closer in. Code still gets produce
-
What Is Baby Cursor?
Ryo Lu built Baby Cursor as a simpler environment where an idea can emerge. It looks exactly like Cursor, and it is a prototyping environment for more ideas.
A shared meme says Cursor is a VS Code fork. The product already sits in someone else's world, and certain things are hard to change. Trying one idea can mean spinning up a backend server, the API, the client, and all the scripts before the experiment itself. Baby Cursor is the simpler environment built so the idea comes first. Parts of it might sit closer to the idea. It might feel really real, and it is still a prototyping environment.
People at Ramp and Notion are building little prototyping environments too. Designers can have fun
-
Is Cursor Only for Programmers?
Ryo Lu says Cursor is not something specific just for programmers or even for coding. The models are themselves general purpose: they have read the whole world, all the public information and knowledge, and all the repositories on GitHub. The tool should fit how a person thinks and the input method they prefer.
Focusing on one problem out of maybe 20, then building the whole concept from that single problem, will not scale and will not be as flexible. Designers love pictures and think in 2D space, sometimes drawing little boxes and sometimes making a high fidelity mock. PMs set constraints in longer text, not a short chat prompt. Coders maybe also build a prototype to set an idea. Pictures,
-
Why Does Caring About Cursor Mean Staying in the Product?
Ryo Lu says caring about Cursor means staying in the product constantly, which is the difference between doing software as a job and building something you really care about. You start from your own pain points, your own habits, and a tool that flows with how you think and makes you happy. The work then grows into solving problems for people like you.
Everyone who started Cursor was a software engineer who loved programming and loved making tools. They needed something that fits AI better and takes advantage of what the models have when the models are put with coding: the right interaction and the right way to do the thing. They built Cursor for themselves, kept going because they care, and