Role: Senior Research Scientist/Engineer (AI Safety)
Location: London, United Kingdom
Job Type: Permanent

The potential of current AI systems is evolving at a rapid pace. While these advancements offer fantastic opportunities, they also bring certain risks like deliberate misuse or intelligent yet misaligned models. Our client is conducting fundamental research on interpretability and behavioral model evaluations which is used to audit real-world models.

You will be responsible for working on safety cases for scheming by building evaluations for deceptive alignment-related properties, such as situational awareness or deceptive reasoning. You will publish the results of your work to the general public or target audience (AI developers or governments)

Requirements

• Extensive LLM experience - the core skills of this role is steering LLMs.
• Software Engineering experience - build a solid foundation in software engineering including APIs, data science, system design, data engineering and front-end front-end development.
• Research experience - design and execution of experiments and then writing up and communicating your findings.
• Generalist - your skills range from LLM fine-tuning to developing frontend interfaces.
• Scientific mindset - experienced proposing and testing alternative hypotheses in new experiments.

Interested? Apply or email harvey.cheadle@signify-tech.com