VisionAi

Built a full-stack AI SaaS product from my room — where the only thing louder than my keyboard was my existential dread. Introducing VisionAI — an AI image generation platform so loaded with features, it might actually be more productive than half the startups on Product Hunt. 🧠 What it does: Generates 2K/4K AI images on command Stripe-powered payments with credit system Free, Pro, and Enterprise plans baked in Community gallery (likes, comments, follows — dopamine simulator) Admin dashboard, user analytics, payment history Role-based JWT auth with email verification No watermark BS (unless you’re broke on Free tier) 🛠️ Tech Stack Breakdown (AKA the sweat and suffering behind the scenes): Frontend React 18 + TypeScript — because vanilla JS is just unpaid suffering Vite — builds faster than my mental breakdowns TailwindCSS — utility-first, braincell-last React Query — async data on roids Radix UI — accessible components that don’t look like 2007 Lucide Icons — open-source icons that slap Backend Node.js + Express — classic battle-tested combo MongoDB + Mongoose — flexible, forgiving, and occasionally cursed JWT — because session cookies are for the weak Multer — for handling file uploads like a boss Stripe — so your users can give you money (which you deserve) Helmet — for when you remember security isn’t optional Infra / DevOps Docker — so you can ship this thing without sacrificing your soul Vercel + Railway — because nothing says “done” like a one-click deploy MongoDB Atlas — cloud-hosted DB without the pain Cloudinary — so your images live somewhere other than /uploads/temp Webhooks, .env configs, production builds — all ready For devs who value time, hate boilerplate, and want to launch fast without writing the same 17 auth flows again: 👉 https://lnkd.in/giTcQYiY No BS. Just clean code, modern stack, and enough features to flex in any investor pitch.