How We Build 2.0
"Sounds like AI" deletes the best thinking first. Using tools to create isn't inhuman.
Every tool since fire had its deniers, and history marks them irrelevant fast.
(Also your tools for detecting this kind of stuff are worse than you think. Pangram flags this note as AI-written. I typed it myself. Point proven.)
This is a copy of the original document I wrote to kick off what would become Spawn.
Mission
Turn software creation from a technical implementation process into a creative exploration process.
I keep telling people the problem isn't "coding is hard" or "development takes too long." The real problem is you have to translate the thing in your head into data models and user flows before you can see any of it, and nobody thinks in data models. You think in "what if it worked like this."
Get rid of that translation step and the technical stuff doesn't need to be abstracted away, it's just not part of the creative act anymore. That's the durable goal, and it's a different business than simplifying development.
Product magic moment
Say you type "I want an app where people can share their favorite hiking trails and rate them." You get back 4 different takes. Maybe one is basically AllTrails but more social, another leans hard into beautiful photo sharing with map overlays. There's a gamified one with achievements for completing trails etc, and then a minimalist one that sneaks in clever stuff like automatic elevation profiles.
Then you mix and match the parts you like, or grab one version and keep going from there.
Way more fun than the usual no-code thing of "okay first drag this database component and configure these fields..."
The part I actually care about is the four, not the idea-to-app trick. Hand someone a single result and they evaluate it like something they ordered, whereas with four in front of them, one usually has a thing they didn't ask for but want. They start going "ooh I like that aspect from version 2..." and the spec never gets written. Somewhere in there it stops being programming made simple and turns into finding out what your app could be.
We shouldn't burn months trying to guess every use case up front either. Nail that one moment and users will show us what they want to build with it.
Watching it get built
Related, and I feel strongly about this: when you hit enter, you should see all 4 variations getting built live, right in front of your eyes.
No spinning wheel for 30 seconds. You'd watch the navigation bar take shape first, content sections flowing in after it, interactive elements showing up and getting styled, data starting to populate.
That does a few things for us. The wait becomes engaging instead of frustrating, with little moments of anticipation and surprise, and you start getting ideas mid-generation ("oh wait, I like how that menu is forming..."). You also actually understand what you're getting, sort of like watching a chef plate your dish.
We could even make the building process itself interactive, things like:
- "Pause, go back to how that list view looked 2 steps ago"
- "Keep that header animation but try something different with the cards"
- "I like where version 3 is going, generate more variations from this point"
All that waiting turns into discovery, and watching the process helps you build up an intuition for how to steer the system toward what you actually want.
So much more alive than waiting around for a fully-formed app to appear. You're watching your idea materialize.
Communicating this concept to other people
Reminds me a bit of how hard Figma was to explain before people had actually used it. "It's like Adobe Illustrator but in the browser" totally missed what made it revolutionary, the multiplayer and components and the rethinking of the whole design workflow.
Maybe the key is to avoid technical comparisons entirely, so rather than positioning against development tools ("it's like X but with AI") we position against creative tools:
"Remember when making images required learning Photoshop? Now with Midjourney you just explore possibilities. We're doing that for apps."
Or focus on the emotional experience: "This isn't about writing better code or building faster - it's a conversation about what could exist, where you watch it take shape in real time."
Some analogies that might work:
- "More like playing with clay than following a blueprint"
- "Think of it as a really talented implementation partner who can instantly prototype any idea you have"
- "If Figma and Midjourney had a baby - exploring a space of possibilities instead of following a build process"
Early product philosophy
Key feelings:
- Inspired
- Empowered
- Motivated
- Surprised
Branding & art directions:
- Play with the idea of emergence/materialization
- Like watching something take shape from nothing
- Could inspire both the name and visual language
- Think of how clay forms on a potter's wheel, or how crystals grow
- Focus on the magical moment of creation
- That split second when an idea becomes real
- Could be reflected in the name, logo animation, interface moments
- Like how Figma's logo is about points connecting, but done in a way that feels artistic rather than technical
- The garden/cultivation metaphor
- You plant an idea and watch it grow and branch
- Less about "building" and more about nurturing
- Could inform both brand voice and visual design
The interface itself could be treated as a canvas/art piece:
- Beautiful transitions as versions materialize
- Thoughtful typography and white space
- Maybe even subtle ambient sound design
- Every interaction should feel like you're shaping something organic
The goal would be to create something that feels more like a creative studio than a development environment - a space that inspires possibilities rather than suggesting implementation.
Ecosystem
- Single auth for any of the apps
- Shared data model (perms?)
- Single payment methods for all of the apps