Launched this week
TextBehind

TextBehind

Create magazine-style covers in seconds

2 followers

TextBehind is a free browser-based tool that uses AI to automatically separate foregrounds from backgrounds. It lets creators place text behind subjects to create professional depth effects for YouTube thumbnails, magazine covers, and social media posts—all in seconds
TextBehind gallery image
TextBehind gallery image
Free Options
Launch tags:Design ToolsSaaSTech
Launch Team / Built With
Famulor AI
Famulor AI
One agent, all channels: phone, web & WhatsApp AI
Promoted

What do you think? …

DARSHAN PRAJAPATH 2441517
Here is the perfect "Maker's Comment" to post immediately after you launch. This comment covers your story, the technical struggles (that we just fixed!), and your "why." It builds trust because it is honest. The "Maker Story" Comment (Copy & Paste) Headline: Why I built this (and how I almost broke it) Hey everyone! 👋 I’m a student developer and the maker of TextBehind. 💡 The Inspiration I love the "magazine cover" aesthetic where text sits elegantly behind a subject. But honestly? I hated opening Photoshop and spending 20 minutes manually masking hair and edges just to make a simple YouTube thumbnail or IG story. I wanted a tool that could do that boring "masking" part in 5 seconds. 🚧 The Problem & The "Chaos" Building this wasn't smooth sailing. The biggest challenge was Performance vs. Stability. The Database Crash: Just yesterday, I was battling constant 500 Internal Server Errors on Vercel. My Supabase database connections were getting exhausted because I wasn't using connection pooling correctly. The Browser Lag: Running AI models client-side (to save server costs) is heavy. My early versions would freeze the browser and trigger "Page Unresponsive" warnings because the math was blocking the main thread. 🛠 How I Solved It Architecture: I stuck to a client-side approach using @imgly for background removal. This ensures user privacy (images don't leave your device) and keeps the tool free since I don't pay for GPU servers. The Fixes: I rewrote the backend logic to handle "Race Conditions" using upsert queries and optimized the AI image resolution to 1024px to stop the UI from freezing. 🚀 The Process I built this v1 using Next.js and Tailwind for speed, Supabase for the backend, and Clerk for auth. It’s simple, fast, and does one thing really well. I’d love for you to try it out and let me know if you hit any bugs (I hope not!).