I’ve been spending a lot of time building with AI, and one thing keeps frustrating me: the models are increasingly multimodal, but the way we interact with them hasn’t really caught up.
We still spend most of our time typing into a text box, attaching screenshots, and writing paragraphs to explain what the AI should look at. Sometimes I just want to circle something on a screenshot, draw a quick diagram, or explain an idea with a voice note.
So I started building Bloccy. It’s a small macOS app built around a canvas for combining text, images, sketches, annotations, and voice notes into a single message.
The important part is that it works with the ChatGPT and Claude conversations I’m already using. I’m not trying to build another AI chat app.
It’s still early, but I’ve been using it myself and figured it was time to share.
