Trust and Safety Software Engineer

Anthropic is seeking a Trust and Safety Software Engineer to join their mission of creating reliable, interpretable, and steerable AI systems. This role is crucial for building safety and oversight mechanisms for AI systems, focusing on monitoring models, preventing misuse, and ensuring user well-being. The position combines technical expertise with safety-conscious development, requiring skills in abuse detection, monitoring systems, and defensive infrastructure.

The ideal candidate will have 3-8+ years of software engineering experience, particularly in integrity, spam, fraud, or abuse detection. They'll work on developing sophisticated monitoring systems, building abuse detection mechanisms, and creating robust multi-layered defenses that operate at scale. The role requires proficiency in SQL, Python, and data analysis tools, along with strong communication skills.

Anthropic offers a collaborative environment where team members work together on large-scale research efforts, focusing on impactful AI development. The company values empirical science approaches and maintains a strong emphasis on communication and research discussions. They provide competitive compensation (£240,000 - £325,000 GBP), flexible working arrangements, and various benefits including visa sponsorship.

Located in San Francisco, Anthropic operates as a public benefit corporation, demonstrating their commitment to societal benefit. They encourage applications from diverse candidates, even those who might not meet every qualification, recognizing the importance of varied perspectives in addressing the social and ethical implications of AI systems. The hybrid work environment requires at least 25% office presence, fostering both flexibility and in-person collaboration.