[Expression of Interest] Research Engineer / Scientist, Alignment - London

Anthropic is hiring a Research Engineer/Scientist for its Alignment Science team in London. The role focuses on designing and running machine learning experiments to understand and steer advanced AI systems, with an emphasis on AI control and alignment stress-testing. You will collaborate with interpretability, fine-tuning, and frontier red-teaming teams to produce research that informs Anthropic's Responsible Scaling Policy.

• Design and run machine learning experiments to evaluate and improve AI safety techniques.
• Test the robustness of safety measures by training language models to subvert interventions.
• Run multi-agent reinforcement learning experiments, such as AI Debate.
• Build tooling to evaluate novel LLM-generated jailbreaks and create evaluation prompts.
• Contribute to research papers, blog posts, and talks.
• Run experiments that inform Anthropic's Responsible Scaling Policy.

• Significant software, ML, or research engineering experience.
• Some experience contributing to empirical AI research projects.
• Familiarity with technical AI safety research.
• Preference for fast-moving collaborative projects over extensive solo efforts.
• Willingness to pick up slack, even beyond job description.
• Care about the impacts of AI.