Frontier Safety Mitigations Research Engineer, DeepMind

DeepMind is seeking a Frontier Safety Mitigations Research Engineer to join their London team. The role involves developing advanced classifiers and data pipelines to identify misuse, as well as creating monitoring systems to detect coordinated harms across user sessions. The engineer will also implement semi-automated response systems to track and manage malicious actors, and evaluate AI systems by developing threat models and testing environments. A strong focus will be on advancing research in automated red-teaming and adversarial robustness. Candidates should possess a Bachelor's degree in Computer Science, Machine Learning, or a related field, along with at least five years of software development experience, particularly in Python. Familiarity with the research-to-deployment pipeline in an AI context is essential, as is experience in a software engineering or research team. While the advert does not specify the employment type, the role is suited for individuals with a strong background in cybersecurity detection.