Hey Product Hunt! I’m Serkan, the maker of Tellwe. 👋

Tellwe takes its name from telve the Turkish word for the coffee grounds left in the cup after drinking, where the fortune is traditionally read. We turned that into Tellwe: a cup with a story to tell.

I grew up around Turkish coffee readings flipping the cup, waiting, then finding birds, roads, hearts, and tiny stories in the grounds. I wanted to turn that ritual into something playful and character-led, rather than another AI chat box.

You upload a photo of your cup. Tellwe checks if it’s readable, finds the symbols, writes a personal 45–50 second reading, creates the voice, and brings the cup character to life with expressions, gestures, synced mouth shapes, and subtitles.

At the end, you get a vertical video you can share while your private cup photo never appears in the public video.

The product runs end-to-end on Vercel using Next.js, Functions, AI Gateway, Workflow, Blob, Sandbox, FFmpeg, and Remotion.

The first version supports English and Turkish.

The hardest part wasn’t making the AI work. It was making a coffee cup feel like a character you actually want to listen to.

I’d love feedback on three things:

1. Does the character feel alive?
2. Does the reading feel fast and engaging?
3. Would you share the finished video?

Thanks for trying Tellwe and for letting a cup tell its story. ☕️

Tellwe

Your coffee cup tells its story.

Your coffee cup tells its story.