In February of 2024, Google released Gemini 1.5, the next generation of their multimodal model Gemini.

This AI research breakthrough is introducing a new era of long-context understanding and multimodal processing capabilities. Gemini 1.5's ability to process up to 1 million tokens allows it to digest and analyze vast amounts of information, from lengthy documents to hours of video, all in a single go.

I collaborated with DeepMind researchers to find effective ways to demonstrate the capabilities of Gemini 1.5.

Video Demos

I helped concept, script and voice a series of tutorial videos exploring use cases leveraging a long context window.

I worked most closely concepting this three.js coding example. I also voiced the Apollo 11 transcript demo video below.