Research Engineer – AI Alignment (London)

About the role

You will design and run machine‑learning experiments to understand and steer the behavior of powerful AI systems, focusing on safety, alignment, and robustness. The role blends scientific inquiry with engineering, collaborating with teams such as Interpretability, Fine‑Tuning, and the Frontier Red Team.

Key responsibilities

• Test robustness of safety techniques by training language models to subvert interventions.
• Run multi‑agent reinforcement learning experiments, e.g., AI Debate.
• Build tooling to efficiently evaluate novel LLM‑generated jailbreaks.
• Write scripts and prompts to generate evaluation questions for safety‑relevant reasoning.
• Contribute ideas, figures, and writing to research papers, blog posts, and talks.
• Run experiments that inform Anthropic’s Responsible Scaling Policy.

Required profile

• Significant software, ML, or research engineering experience.
• Experience contributing to empirical AI research projects.
• Familiarity with technical AI safety research.
• Preference for fast‑moving collaborative projects.
• Willingness to pick up tasks outside the job description.
• Commitment to the impacts of AI.

Required skills

• Python
• Large language models (LLMs)
• Reinforcement learning
• Kubernetes
• Working with complex shared codebases

What we offer

• Annual salary £260,000–£370,000 GBP.
• Competitive benefits including equity donation matching, generous vacation and parental leave, flexible working hours, and office space.