Research Scientist, Human Data for Robotic Intelligence

Google LLC; Greater London; On-site; GBP 70,000 - 110,000; Full time

Job summary

Google DeepMind in London seeks a Research Scientist to advance human-data-driven robotics and to develop scalable robotic foundation models. You will design and evaluate large multimodal models, run experiments, and collaborate with a broad team across AI disciplines to translate research into real-world robotics capabilities.

The role emphasizes incorporating human action data, working with simulators and real robots, and publishing findings with academic partners.

Qualifications

• PhD in CS or a related field, or equivalent practical experience.
• 2 years of experience with reinforcement learning and imitation learning.
• Experience with multimodal models, including vision-language and video.
• Experience with multimodal generative modeling, training and inference.

Responsibilities

• Design, implement, train and evaluate large models and algorithms for robotic agents, making breakthroughs and unlocking new robot capabilities.
• Write software to implement research ideas and iterate quickly, working effectively with a large collaborative team with fast-paced agendas to meet ambitious research goals.
• Develop methodologies and design and conduct experiments for incorporating scalable data sources, especially human data with or without capture devices into our robotics foundation models.
• Leverage broader expertise to participate in a wide variety of research, including learning from simulation, reinforcement learning, learning from demonstrations, vision-language-action models, transformers, video generation, robot control, humanoid robots and more.
• Report and present research findings clearly and efficiently both internally and externally.

Skills

Reinforcement learning

Imitation learning

Multimodal models

Vision-language

Generative modeling

Education

PhD in Computer Science or related field

Job description

Google DeepMind in London seeks a Research Scientist to advance human-data-driven robotics and to develop scalable robotic foundation models. You will design and evaluate large multimodal models, run experiments, and collaborate with a broad team across AI disciplines to translate research into real-world robotics capabilities.

The role emphasizes incorporating human action data, working with simulators and real robots, and publishing findings with academic partners.