Enforcement Analyst Agentic Alignment
AI safety and research company building reliable, interpretable, and steerable AI systems, including the Claude product family and developer platform.
Funding history
About Anthropic
Anthropic PBC develops frontier AI systems and deploys them through Claude products and the Claude Platform, with a stated focus on safety, interpretability, and steerability.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will investigate harmful and unexpected model behavior using transcripts, tool-use traces, code, logs, and data. You will triage cases, make and document enforcement decisions, assess cyber-related activity, write evidence-based findings, and improve review workflows. You may encounter explicit sexual, violent, or psychologically disturbing content and participate in on-call responsibilities.
Requirements
- Ability to read and understand Python, JavaScript, or shell scripts
- Background in security, cyber threat intelligence, or a similar technical field
- Experience in trust and safety, enforcement, investigations, or a related field
- Experience conducting structured investigations
- Experience supporting formal enforcement actions, case documentation, evidence packaging, and escalation coordination
- Clear writing and stakeholder communication
- Judgment about escalation and resilience when working with difficult content
- Bachelor’s degree or equivalent education, training, or experience
Responsibilities
- Review and investigate potential agentic-misalignment cases
- Triage cases by severity and urgency
- Operationalize scalable review pipelines for agentic cases
- Make and document Usage Policy enforcement decisions
- Assess cyber-related agentic activity and real-world harm potential
- Write evidence-based case reports for varied audiences
- Identify case trends and provide findings to policy, detection, and research teams
- Maintain enforcement documentation, review logic, and precedents
- Collaborate with Threat Intelligence, Research, Engineering, and Policy partners
- Use AI systems to summarize transcripts, analyze code, query data, and build lightweight tooling
- Contribute to reports on agentic harm trends and enforcement outcomes
Benefits
- Optional equity donation matching
- Generous vacation
- Parental leave
- Flexible working hours
