Inspiration

We’ve all been there: you have a great idea for a post, but you spend 40 minutes scrolling through stock photo sites or fighting with prompt engineering, only to end up with a "perfect" image that has misspelled text. I built Posterly because creators shouldn't have to choose between the speed of AI and the control of a designer.

What it does

Posterly is a lightning-fast design tool that generates unique, high-quality posters from your ideas and—crucially—keeps the text fully editable. It transforms a simple thought into a platform-ready visual for Twitter, LinkedIn, or Instagram in seconds, allowing you to tweak the copy until it’s perfect.

How I built it

The Brain: Leveraged modern AI image generation models (Gemini 3) to create the visual foundation. The Canvas: Built a custom dynamic overlay system that maps text layers on top of AI-generated backgrounds. The Stack: [Insert your tech stack, e.g., Next.js, Tailwind, and Gemini API].

Challenges I ran into

The biggest hurdle was the "Layer Logic." Standard AI generates flat images. I had to develop a workflow that treats the background and the typography as separate entities, ensuring that when a user edits a word, it doesn't break the aesthetic integrity of the generated design.

Accomplishments that I am proud of

I’m most proud of the UI fluidity. We managed to reduce the "idea-to-export" pipeline from nearly an hour of manual searching to under 60 seconds of creative flow.

What I learned

This project taught me that "AI-assisted" is often better than "AI-automated." Users don't want the machine to do everything; they want the machine to do the heavy lifting so they can focus on the final creative touch.

What's next for Posterly

The vision is to implement Smart Layouts—where the AI suggests different typography styles based on the sentiment of your text—and direct API integrations to post your finished poster to social media with one click.

Built With

Share this project:

Updates