Senior Research Scientist, Gemini Omni, DeepMind

About the job

We research and develop machine learning models for billions of Google users.

Artificial intelligence will be one of humanity’s most transformative inventions. At Google DeepMind, we are a pioneering AI lab with exceptional interdisciplinary teams focused on advancing AI development to solve complex global challenges and accelerate high-quality product innovation for billions of users. We use our technologies for widespread public benefit and scientific discovery, ensuring safety and ethics are always our highest priority.

Minimum qualifications:

• PhD in Computer Science, a related field, or equivalent practical experience.
• 3 years of experience in development with JAX, PyTorch, or TensorFlow.
• 3 years of experience with machine learning and machine learning algorithms.
• 3 years of experience with Generative Artificial Intelligence (GenAI) techniques (e.g., Large Language Models, Multi-Modal, Large Vision Models) or with GenAI-related concepts (language modeling, computer vision).
• 2 years of experience leading a research agenda.
• Experience in academic research within machine learning, publications, or research in related fields.

Preferred qualifications:

• 2 years of coding experience.
• 1 year of experience leading research efforts and influencing other researchers.

Responsibilities

• Drive foundational research in next-generation generative modeling (e.g., autoregressive architectures, diffusion models) to pioneer breakthroughs across multimodal generation, including image, video, and audio synthesis.
• Formulate novel scientific methodologies and conduct deep literature reviews to solve complex, open-ended AI challenges, exercising independent judgment to balance immediate project milestones with long-term frontier research.
• Pioneer model optimization and efficient inference strategies to significantly reduce compute overhead, optimize latency, and scale large multimodal systems across high-performance infrastructure.
• Advance post-training and capability scaling using reinforcement learning to enhance model alignment, reasoning, and multi-turn generation quality across modalities.
• Apply rigorous engineering and experimental practices to design robust benchmarks, measure real-world performance, and systematically validate the scientific and practical impact of research findings.