Research Scientist, World Models, DeepMind

  • Full Time
  • London

Google

Research Scientist, World Models, DeepMind Overview

Company Name Google
Job Role Research Scientist, World Models, DeepMind
Qualifications Bachelor’s
Category IT Jobs
Job Type Full Time
Location London

We are seeking a Research Scientist to join our team focused on developing multimodal generative models and advancing the field of world modeling research. This position emphasizes the application of diffusion models for real-time video generation and involves working closely with our core modeling team to create capable and interactive models.

In this role, you will collaborate with researchers and engineers to establish effective feedback loops between modeling and data initiatives. Your work will encompass significant research, road-mapping, and engineering tasks aimed at delivering innovative world models.

Responsibilities

  • Conduct research and implement innovative architectures to create advanced generative models with real-time capabilities.
  • Design, assess, and generate datasets to uncover new functionalities of models.
  • Collaborate across various capability workstreams to identify and scale effective training methodologies.
  • Engage with the training stack to integrate new generative and world model features.
  • Stay updated with the latest research and developments in diffusion models and generative video architectures.

Requirements

  • A PhD in Computer Science, Machine Learning, Statistics, Mathematics, or a related technical discipline, or equivalent practical experience.
  • At least 2 years of experience in publishing research at machine learning conferences or in releasing software models, whether open or closed-source.
  • A minimum of 2 years of experience working with diffusion models, multimodal generative models, or generative video models.
  • Preferred: 3 years of experience in leading AI research labs, particularly in pre-training or post-training teams.
  • Experience in writing TPU/GPU kernels (such as JAX or CUDA) to enhance model performance and enable real-time inference.
  • Experience in designing, training, and scaling generative pixel or real-time video architectures.
  • A proven track record of collaborating across functions to deliver models.

Benefits

  • Opportunity to work in a pioneering AI lab focused on solving complex global challenges.
  • Access to diverse learning opportunities and various career pathways.
  • Engagement in significant research and engineering projects that contribute to public benefit and scientific discovery.

Degree Requirement: Bachelor’s

Visa Sponsorship May be

To apply for this job please visit www.google.com.