2 of 2 Remote/Hybrid Black-Box Testing Jobs in England

confidence intervals and hypothesis testing, power analysis, experimental design, causal inference, and Bayesian modeling. Experience designing evaluation methodologies - including black-box and counterfactual approaches - and the test datasets behind them, with good judgment on realism and validity limits. Practical familiarity with evaluating AI systems … into testable criteria. First-hand experience with how modern agentic AI systems are built and their behavior emerges Experience with explainability for black-box systems (surrogate modeling, behavioral analysis, etc) This role isn’t for you if... You've only worked in academic or research settings ...

capable model attempting to cause harm is an empirical question that remains largely untested. The Control Red Team at the company is stress‐testing control monitors (including from GDM and Anthropic) – finding attacks that beat monitors and attempting to design experiments to measure real‐world risk. … that generate the evidence, writing arguments, and arguing them out with the rest of the team. We intend to publish this work. 2. Testing: Running evaluations of frontier labs' monitors, and reporting the implications. Turning our research into concrete assessments of real systems: threat modelling ...