Senior Research Scientist — Post-Training Consumer Agents

DeepMind Technologies Limited; San Francisco (CA); On-site; USD 262,000 - 364,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

Life insurance

Disability insurance

401(k) match

Paid time off

Sick leave

Maternity leave

Baby bonding

Holidays

Fair Chance Act

Job summary

Google DeepMind is seeking a Research Engineer/Research Scientist to work on Agent Post-training and Personalization, building autonomous consumer agents through reinforcement learning and targeted data creation.

You will develop datasets and evaluation frameworks to test long-horizon reasoning and measure agent reliability across complex workflows. Join a leading AI lab driving safe, high-impact AI innovation.

Qualifications

• Bachelor's degree or equivalent practical experience.
• 8 years of software development experience.
• 5 years in design, architecture and launching software products.
• Experience with large language models, RL and deep learning.
• Master's or PhD in Engineering/CS or related field preferred.
• 8 years of data structures/algorithms experience.
• 5 years in technical leadership with project teams.
• 3 years in matrixed cross-functional environments.

Responsibilities

• Specialize in core modeling and post-training research for agent post-training.
• Design and run ablation studies to validate hypotheses.
• Communicate findings and progress to the team with visualizations.
• Develop research infrastructure for data analysis and model evaluations.
• Collaborate with other researchers and engineers for feedback loops.

Skills

LLMs

Reinforcement learning

Deep learning

Software development

Architectural design

Education

Bachelor's degree

Master's or PhD in Engineering or CS

Job description

Google DeepMind is seeking a Research Engineer/Research Scientist to work on Agent Post-training and Personalization, building autonomous consumer agents through reinforcement learning and targeted data creation.

You will develop datasets and evaluation frameworks to test long-horizon reasoning and measure agent reliability across complex workflows. Join a leading AI lab driving safe, high-impact AI innovation.