Hylo raised $1.5M and shipped fast. Maybe too fast. xTokens, LSTs, their own yield-bearing stablecoin, guys were locked in af. But every module looked like a different product. Land in xTokens, jump to staking, and you'd swear you switched apps entirely. I came in and rebuilt it into one system. 60+ components, a real design system, shadcn/ui implemented directly in code, so the dev team ships faster and nothing drifts between Figma and prod. Scattered MVP → something that reads like it belongs at the top of the Solana stack.

Not all design feedback should carry the same weight. I sort it into risk, constraint, or preference. Risk and constraint can reopen a decision. Preference needs an explicit tradeoff and a named decision owner. In your next review, label the input before debating it.

Never stop hustling

Brainstorming of AI use cases turned into @joinvitrine pitch to @OobleStudio :D

I’m not drinking coffee, but I’ll take it as an advice

Shit, that was close

“Nothing here yet” is a description, not a first step. Tell me what belongs here, how to add it, and what I’ll be able to do once it’s there. Try that on one empty screen in your product.

Cooking 👩‍🍳

Craft is what matters most these days cause anything can be vibe-coded It's a ton of todo apps, yet thousands of people still use @culturedcode It's a ton of note taking apps, yet thousands still use @craftdocsapp If the logic is bones, the UX is meat, the UI is skin, then a well-crafted nitpicks are the cloth. And technically the cheap dirty T-Shirt and the brand new Lacoste polo solve the same issue, but you know...

Good luck guys! If you wanna work with $1M project, help them early yeah? 🙃