Enforcement Analyst, Agentic Alignment

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

Benefits offered by this job

Competitive compensation and benefits

Equity donation matching (optional)

Parental leave

Flexible working hours

Office workspace for collaboration

Visa sponsorship with immigration law

Job summary

Anthropic is hiring an Enforcement Analyst focused on Agentic Alignment to identify, investigate, and enforce decisions on harmful or unexpected agentic model behavior. You will triage cases, document decisions, and collaborate with policy, research, engineering, and threat intelligence teams while using Claude to accelerate investigations.

You will operationalize scalable review workflows, summarize transcripts, analyze code, and contribute to reports on agentic harm trends and enforcement

Qualifications

• Background in security, cyber threat intelligence, or a technical field enabling assessment of cyber harm and severity.
• Experience in trust & safety, enforcement, investigations, or related fields with defensible decisions and clear findings.
• Experience conducting structured investigations from first signal to final finding and escalation coordination.

Responsibilities

• Review and investigate potential agentic misalignment using transcripts, tool-use traces, and code.
• Triage cases by severity and urgency and communicate through appropriate channels.
• Operationalize review pipelines for agentic cases and translate detection into scalable workflows.
• Make enforcement decisions under the Usage Policy with documented reasoning and precedents.
• Assess cyber-related agentic activity to evaluate real-world harm potential.
• Write clear, evidence-based case write-ups for diverse audiences.
• Identify patterns across cases and feed findings back to policy, detection, and research teams.

Skills

Security background

Trust & safety

Investigation experience

Written communication

Enforcement decisions

Cross-functional collaboration

Education

Bachelor's degree or equivalent

Tools

Claude

Job description

Boston, MA; San Francisco, CA | Seattle, WA | New York City, NY | Washington, DC

Uses AI assistants (Claude) to accelerate investigations and analyze agent behavior; directly involves investigating autonomous AI behavior.

About the Role

Anthropic is hiring an Enforcement Analyst focused on Agentic Alignment to identify, investigate, and make enforcement decisions about harmful or unexpected agentic model behavior. The role involves triaging cases, documenting decisions, collaborating with policy, research, engineering, and threat intelligence teams, and using AI systems to accelerate investigations.

Job Description

Role

As an Enforcement Analyst (Agentic Alignment) you will identify and investigate harmful or unexpected agentic behavior by models, triage cases by severity, make enforcement decisions under the Usage Policy, and operationalize scalable review workflows. You will collaborate across Policy, Threat Intel, Engineering, Research, and Transparency teams and contribute findings back to improve detection, policy, and model behavior.

Key Responsibilities

• Review and investigate potential instances of agentic misalignment using transcripts, tool-use traces, and code.
• Triage cases by severity and urgency and communicate through appropriate channels.
• Operationalize review pipelines for agentic cases and translate detection outputs into structured, scalable workflows.
• Make enforcement decisions under the Usage Policy, applying consistent reasoning and documenting decisions and precedents.
• Assess cyber-related agentic activity (scanning, exploitation attempts, credential misuse, malicious code) to evaluate real-world harm potential.
• Write clear, evidence-based case write-ups for audiences from enforcement peers to senior leadership.
• Identify patterns and trends across cases and feed findings back to policy, detection, and research teams to improve outcomes.
• Maintain rigorous documentation of enforcement decisions, review logic, and precedents.
• Collaborate with Threat Intelligence, Research, Engineering, and Policy partners to ensure detection coverage keeps pace with evolving agentic misuse.
• Use AI systems (including Claude) to accelerate investigations: summarizing transcripts, analyzing code, querying data, and building lightweight tooling.
• Contribute to regular reports on agentic harm trends and enforcement outcomes.

Requirements

• Background in security, cyber threat intelligence, or a similar technical field enabling assessment of cyber harm and severity.
• Experience in trust & safety, enforcement, investigations, or related fields, translating ambiguous policy questions into defensible enforcement decisions and clear written findings.
• Experience conducting structured investigations from first signal to final finding and judging when evidence is sufficient to act.
• Experience supporting or contributing to formal enforcement actions, including case documentation, evidence packaging, and escalation coordination.
• Strong written communication and ability to present complex findings to Legal, Policy, Engineering, and non-technical stakeholders.
• Sound judgment about escalation and resilience to work with difficult or explicit content. Note: on-call responsibilities across Policy and Enforcement teams.

Preferred qualifications

• Demonstrated use of AI systems to accelerate work and a willingness to continue innovating with them.
• Comfortable working with large volumes of data (long transcripts, logs, traces, multi-table datasets) and surfacing behavioral patterns.
• Curiosity about AI agent behavior and experience investigating novel behaviors without established playbooks.
• Familiarity with AI safety and alignment and the risks of increasingly autonomous AI systems.

Logistics

• Annual salary range: $245,000 - $285,000 USD.
• Minimum education: Bachelor’s degree or equivalent combination of education, training, and/or experience.
• Location-based hybrid policy: staff are expected to be in one of Anthropic’s offices at least 25% of the time.
• Visa sponsorship: Anthropic will make reasonable efforts to sponsor visas and retains immigration legal support.

Skills

Investigations Case Triage Threat Intelligence Cybersecurity Policy Interpretation Enforcement Decision-Making Written Communication Cross-functional Collaboration Documentation Judgment Data Analysis Pattern Recognition On-call Operations Resilience

Experience Level

Senior

USD 245,000 - 285,000/year

• Competitive compensation and benefits
• Optional equity donation matching
• Parental leave
• Flexible working hours
• Office workspace for collaboration
• Visa sponsorship (with immigration lawyer support)