This project is a rapid spatial interaction experiment utilizing Gemini Omni’s native multimodal capabilities to bridge physical gesture and generative media. Starting with a raw video clip, I used Omni’s conversational editing framework to seamlessly inject a responsive, lightweight digital canvas directly between my hands, displaying a live render of my portfolio website. By bypassing traditional post-production timelines and leveraging an AI model that natively understands physics and spatial geometry, this experiment showcases a new workflow for design technologists: treating video as an iterative, natural-language prompt to rapidly prototype future spatial computing interactions.