Software Engineer, Safeguards
AnthropicSan Francisco, California, United States
Posted October 14, 2025
320000 - 425000 USD year
Full Time
About the Company
Anthropic is an AI research company focused on developing reliable, interpretable, and steerable AI systems. Their flagship product, Claude, is an AI assistant designed for tasks of any scale. The company's research encompasses natural language processing, human feedback mechanisms, scaling laws, reinforcement learning, code generation, and interpretability, setting them apart with a commitment to transparency and control in AI development.
Job Description
Anthropic is seeking software engineers to build safety and oversight mechanisms for their AI systems. The role involves monitoring models, preventing misuse, ensuring user well-being, and upholding safety and transparency principles. Engineers will develop monitoring systems, detect abuse, build defense mechanisms, and analyze user reports.
Requirements
- Develop monitoring systems to detect unwanted behaviors.
- Build abuse detection mechanisms and infrastructure.
- Analyze user reports of inappropriate content or accounts.
Benefits
- Competitive compensation
- Optional equity donation matching
- Flexible working hours
- Lovely office space