Research Scientist

About the role

In this role, you will conduct original technical research in AI safety to build the foundation for safe, trustworthy AI deployment at Faculty. You will design and execute impactful experiments, study model behavior, and develop robust, reproducible tooling that advances the state of the art. You will work alongside Senior and Lead Research Scientists, bridging scientific discovery with practical application for external partners and national security institutes. This is your chance to hone your expertise, shape emerging safety standards, and solve some of the most critical challenges in modern AI. You will own the end-to-end research lifecycle, translating complex technical insights into actionable recommendations for frontier labs and government bodies. You will be empowered to envision the most powerful applications of AI and to make them happen within a human-centric framework.

Key facts

What you'll do

• Executing technical research in AI safety by designing and running experiments to study and steer model behavior across diverse scenarios and risk domains.
• Applying white-box and black-box methodologies, focusing on interpretability, steering, and behavioral studies of transformer-based LLMs to uncover emergent properties.
• Developing high-quality, modular, and reproducible research code in Python using deep learning frameworks like PyTorch to ensure scalability and reliability.
• Building and maintaining research infrastructure and technical tooling to accelerate scientific discovery and streamline experimentation workflows.
• Documenting and communicating experimental findings through high-quality academic papers, technical reports, and presentations for both technical and non-technical audiences.
• Collaborating with senior scientists to identify literature gaps, refine hypotheses, and define research scope aligned with real-world client and regulatory needs.
• Translating complex technical insights for diverse audiences, including external frontier labs, government bodies, and safety institutes, to drive informed decision-making.
• Partnering with cross-functional teams to integrate research outcomes into practical solutions that enhance trust, safety, and performance in deployed AI systems.
• Contributing to the development of industry standards and best practices for AI safety through active engagement with the broader research community.
• Driving innovation by exploring novel techniques and edge cases that challenge existing assumptions and expand the boundaries of safe AI deployment.

Requirements

• Solid background in machine learning fundamentals and transformer architectures (specifically LLMs) with a proven track record of academic or applied research.
• A practical understanding of white-box and black-box interpretability techniques, such as steering vectors or mechanistic interpretability, and their limitations in real deployments.
• Demonstrated capability to take structured technical problems and formulate hypotheses that can be tested through rigorous experimentation.
• Strong programming skills in Python and deep learning frameworks like PyTorch, with a track record of writing clean, modular research code that is maintainable and extensible.
• Excellent communication, enabling you to translate complex technical concepts and experimental results into clear reports, visualizations, and presentations.
• Curiosity and analytical rigour, combined with exposure to core AI safety areas such as robustness, evaluations, or uncertainty calibration in high-stakes environments.
• Experience working with sensitive or high-impact domains such as government, finance, life sciences, or defence, and an understanding of the associated constraints and ethical considerations.
• Commitment to responsible AI practices, including thorough documentation, reproducibility, and proactive risk assessment throughout the research lifecycle.

Practical notes

Our recruitment ethos

At Faculty, we aim to grow the best team - not the most similar one. We know that diversity of individuals fosters diversity of thought, and that strengthens our principle of seeking truth. Diverse teams deliver better work, relevant to the world in which we live, and we are united by a deep intellectual curiosity and a desire to use our abilities for measurable positive impact. We strongly encourage applications from people of all backgrounds, ethnicities, genders, religions, and sexual orientations.

Some of our standout benefits include an Unlimited Annual Leave Policy, Private healthcare and dental, Enhanced parental leave, Family-Friendly Flexibility & Flexible working, Sanctus Coaching, and Hybrid Working. If you do not feel you meet all the requirements, but are excited by the role and know you bring some key strengths, please do not hesitate in applying as you might be right for this role, or other roles. We are open to conversations about part-time hours and welcome candidates who are passionate about contributing to cutting edge AI safety research even if their exact background varies from the core requirements.