Research Engineer, Frontier Safety Mitigations

London, UK

• In this role, you'll de-risk model launches by defending against misuse and building safety mitigations.
• Build classifiers and data pipelines to detect misuse from evaluation to rapid model iteration.
• Develop cross-context monitoring systems to detect coordinated harms across user sessions.
• Implement data-driven response systems to detect and track persistent malicious actors.
• Evaluate agentic AI systems by developing threat models and deploying robust mitigations.

Google DeepMind is a frontier AI research and product company, with teams working on alignment, policy, and security. We post specific opportunities at Google DeepMind that we think may be high impact. We do not necessarily recommend working at other positions at Google DeepMind. You can read concerns about doing harm by working at a frontier AI company in our career review on the topic.