Senior AI Safety Research Scientist - High-Impact Evaluations

Fdetimes; Greater London; Hybrid; GBP 110,000 - 150,000; Full time

Job summary

Faculty’s central R&D team is seeking a Senior Research Scientist to lead the development of cutting-edge safety evaluations. You will quantify AI risks in high‑impact domains like CBRN and Cyber, driving original research that advances evaluation methodology while collaborating with delivery teams building evaluations and red‑teaming for frontier labs.

You’ll contribute to scientific leadership, publish results, and engage with the global research community and government stakeholders.

Qualifications

• Own research end-to-end—from problem framing to publication.
• Experience designing AI evaluations or benchmarks.
• Expertise in red-teaming, adversarial testing, jailbreaking, and indirect prompt injection.
• Strong foundation in experimental design and uncertainty quantification (Bayesian methods).
• Deep knowledge of language models, generative AI architectures, training methodologies, and safety mitigations.
• Solid Python proficiency and ability to build robust, reproducible research.

Responsibilities

• Lead the development of novel safety evaluations in high‑impact domains like CBRN and Cyber to quantify risks.
• Conduct original technical research in AI safety evaluation methods from concept to publication.
• Shape the R&D agenda by identifying opportunities to advance safety evaluation methodology.
• Contribute thought leadership to client delivery projects and red‑teaming for frontier labs.
• Represent Faculty’s scientific leadership through external engagement with the global research community and stakeholders.

Skills

Research leadership

AI safety evaluations

Red-teaming

Prompt injection

Bayesian methods

Language models

Python

Experimental design

Tools

Python

Job description

Faculty’s central R&D team is seeking a Senior Research Scientist to lead the development of cutting-edge safety evaluations. You will quantify AI risks in high‑impact domains like CBRN and Cyber, driving original research that advances evaluation methodology while collaborating with delivery teams building evaluations and red‑teaming for frontier labs.

You’ll contribute to scientific leadership, publish results, and engage with the global research community and government stakeholders.