[Expression of Interest] Research Engineer / Scientist, Alignment - London
Anthropic London £260,000 – £370,000 pa
Conducts exploratory machine learning experiments on AI safety and alignment, including AI control, misalignment stress-testing, multi-agent reinforcement learning, and jailbreak evaluation tooling. Works closely with Interpretability, Fine-Tuning, and Frontier Red Team, contributing to research papers and Anthropic's Responsible Scaling Policy. Interviews are conducted in Python.