Security Risk Engineer: AI Risk Quantification & Decisions

Anthropic; New York (NY); On-site; USD 120,000 - 180,000; Full time

Benefits offered by this job

Commuter benefits

Annual education stipend

Home office stipend

Job summary

Anthropic is seeking a Security Risk Analyst to lead quantitative and qualitative risk assessments. You will work with Security Engineering to model and measure risk, using FAIR, calibrated estimation, and simulation techniques.

You will translate complex risk findings into actionable decisions for leadership, while building scalable risk analytics for the organization. The role requires strong collaboration, an ability to manage ambiguity, and a deep commitment to AI safety.

Qualifications

• Hands-on FAIR-style quantification experience including scenario decomposition and Monte Carlo simulation.
• Bachelor’s degree or equivalent; experience in security risk analysis preferred.
• Background in security engineering, detection, threat intelligence, or decision science.

Responsibilities

• Lead quantitative analysis of top security risk scenarios using FAIR, calibrated estimation, and simulations.
• Partner with Security Engineering to assess threat scenarios and control effectiveness.
• Frame risk decisions with clear recommendations and uncertainty, ready for leadership review.
• Scale risk analysis into reusable methods, templates, and training for partner teams.

Skills

FAIR quantification

Monte Carlo

Python

R

Calibrated estimation

Security engineering

Decision science

AI safety

Risk analysis

LLM tooling

Education

Bachelor's degree

Tools

Python

R

Job description

Anthropic is seeking a Security Risk Analyst to lead quantitative and qualitative risk assessments. You will work with Security Engineering to model and measure risk, using FAIR, calibrated estimation, and simulation techniques.

You will translate complex risk findings into actionable decisions for leadership, while building scalable risk analytics for the organization. The role requires strong collaboration, an ability to manage ambiguity, and a deep commitment to AI safety.