I am a Director and Research Scientist at Google DeepMind, co-leading the research behind Gemini’s video capabilities. My team addresses the full scope of video understanding, from exocentric analysis (YouTube, Vertex AI) to egocentric real-time reasoning (Gemini Live). Our approach to native multimodal understanding yields state-of-the-art performance (Gemini 3.0 blog and the companion vision blog).
I aim to solve video understanding as an essential component of multimodal intelligence, serving both as a catalyst for new capabilities and an enabler for multimodal generation. Prior to DeepMind, I worked on Generative AI and large-scale vision. I received my PhD from ETH Zurich which was focused on efficient data selection and clustering algorithms.