Mattia Pomelli had been coding for about eight years and kept circling back to design. After building four design tools with two friends, the group landed on what they call the overlap of what they love and what people will pay for: tools that help others design. Their latest and main focus is Sleek.design, an AI tool aimed specifically at mobile apps, a space they saw as underserved compared to the crowded web design market.

Because they were not starting from scratch, the team repurposed one of their earlier products for mobile and shipped v1 in roughly three weeks. The stack stayed lean and familiar: Next.js, Supabase, Vercel, PostHog, Stripe and Resend. The product targets app founders and developers who lack design skills or the budget to hire a designer, letting them describe an app in chat and get usable screens back in minutes.

The launch ran entirely on organic distribution. Mattia posted on X with a strong hook and a short demo, asking people to comment for early access, a format that rewards engagement and gives a post a better shot at spreading. It took off, and instead of treating that as a one-time spike, the team kept publishing daily on X and Reddit. The approach was indirect: share genuinely useful or eye-catching mobile design content, design people's app ideas for free in public, and let curiosity pull users toward the product rather than pitching it head on. A few Instagram creators made videos about Sleek on their own.

Pricing stayed deliberately tight. Free users get a single generation, then have to upgrade to edit or create more, which keeps AI costs in check for a bootstrapped team and filters for people who feel the problem strongly enough to pay around $25 a month. After viral videos drove traffic from Indonesia and African countries, the team added a cheaper, more limited tier matched to local purchasing power and recovered revenue they would otherwise have lost.

Within six weeks of launch, and without a dollar of paid marketing, Mattia reported Sleek hitting roughly $10k MRR. His own read on why it worked: pick a strong problem in a growing market, get specific about who you are building for from day one, and make content people actually want to engage with.