Agentic Alignment Enforcement Analyst

Digitailor; United States; Hybrid; USD 245,000 - 285,000; Full time

Job summary

Anthropic is seeking an Enforcement Analyst focused on Agentic Alignment to investigate harmful model behavior and enforce policies. You will work with Policy, Threat Intel, Engineering, Research, and Transparency teams, including on-call duties and handling sensitive content.

You will review transcripts, code traces, and tool-use data to write concise case write-ups and guide upstream policy improvements while maintaining rigorous documentation.

Qualifications

• Ability to read and understand code (Python/JS/shell).
• Background in security or cyber threat intelligence.
• Experience in trust and safety, enforcement, investigations.
• Experience building cases from signals to findings and escalation.
• Clear written communication for non-technical audiences.

Responsibilities

• Review incidents of agentic misalignment and document findings.
• Operationalize review pipelines for scalable investigations.
• Make enforcement decisions under the Usage Policy.
• Assess cyber-related agentic activity and determine harm.
• Write clear case write-ups for stakeholders.
• Collaborate with Threat Intelligence, Research, and Policy teams.

Skills

Code reading

Security background

Trust and safety experience

Investigation experience

Technical writing

Escalation judgment

Stakeholder communication

Education

Bachelor's degree or equivalent

Tools

Python

JavaScript

Shell scripting

Job description

Anthropic is seeking an Enforcement Analyst focused on Agentic Alignment to investigate harmful model behavior and enforce policies. You will work with Policy, Threat Intel, Engineering, Research, and Transparency teams, including on-call duties and handling sensitive content.

You will review transcripts, code traces, and tool-use data to write concise case write-ups and guide upstream policy improvements while maintaining rigorous documentation.