Research Scientist, Safety Oversight

Highlights

• Turn large-scale production traffic into intelligence on deployed AI safety.
• Build cross-context monitoring to detect coordinated harms and attack vectors.
• Measure real-world efficacy of safety training and mitigations for Gemini models.

What they offer

• 20% bonus target
• equity
• benefits

Ideal candidate

A safety-minded researcher who cares about proactive oversight of frontier models and thrives turning real-world deployment data into actionable safety insights at scale.

Requirements

• PhD in Computer Science, a related field, or equivalent practical experience
• 3 years of experience building and shipping technical products
• Experience in generative AI and Large Language Models (LLM)
• 3 years of experience developing code, running experiments and analyses collaboratively with coding agents
• Experience building large-scale, highly parallelised data pipelines, working on data quality, automated evaluation design and simple statistical modeling
• Ability to approach new research questions and implement technical solutions for them at scale
• Ability to use AI every day to build and push the frontier of model capabilities