Staff+ Security Engineer, Risk Engineering
San Francisco, CA | New York City, NY | Seattle, WA
Posted 16d ago
About the job
Anthropic is building reliable, interpretable, and steerable AI systems to ensure AI is safe and beneficial. The Security Risk team is rebuilding risk management as an engineering function, leveraging automation and AI-native platforms. This role involves identifying, prioritizing, and treating the company's most critical security risks across a broad spectrum of systems, from authorization to cryptography. You will help define the future of risk management for frontier AI companies, with a direct impact on CISO-level decisions and the opportunity to build the underlying AI-native platform. The work is largely greenfield, allowing you to shape the architecture and discipline.
Responsibilities
- Own and drive complex security risk problems end-to-end with minimal oversight, translating ambiguous signals into a defensible view of severity and likelihood, and managing escalation, treatment, and remediation.
- Build systems for measurable risk and scalable risk work, including quantification tooling, automated intake and triage, and observability for partner teams.
- Act as a calibrated technical peer to Security Engineering, pressure testing architectures, translating findings into prioritized remediation roadmaps, and making investment cases for security fixes.
- Mentor engineers and risk practitioners, fostering a risk engineering culture where teams own their risks and the team provides visibility and judgment.
- Work across Anthropic's security landscape (identity, secrets management, developer security, supply chain, infrastructure, secure frameworks) and build context to reason about risk credibly.
- Go deep into system design and failure modes to ensure assessments and treatment plans reflect engineering reality.
- Identify systematic risks through threat modeling and structured assessment, driving severity calibration and escalation conversations with clear recommendations.
- Contribute to quantitative risk work using methods like calibrated estimation and Monte Carlo simulation.
- Design and build AI-native risk tooling using Claude for risk classification, triage augmentation, and continuous sensing of risk landscape changes.
- Create dashboards and data pipelines for real-time risk posture visibility for engineering and product teams.
- Partner with Security Engineering and risk owners to design remediation roadmaps with clear sequencing, ownership, and investment.
- Measure outcomes by decisions made and risk reduced, rather than just activity.
Requirements
- At least 8 years of software engineering or security engineering experience, including leading and remediating complex security risks independently.
- Bachelor's degree in a related field or equivalent experience.
- Strong programming skills in Python or at least one systems language (Go, Rust, C/C++).
- Broad knowledge across core security engineering domains with depth in at least one (identity/secrets management, developer security/supply chain, infrastructure/cloud security, secure frameworks).
- Calibrated risk judgment to assess severity and likelihood of ambiguous problems and adapt based on evidence.
- Experience leading cross-functional security initiatives and navigating complex organizational dynamics.
- Outstanding communication skills, effectively translating technical concepts across all organizational levels.
- Proven track record of bringing clarity and ownership to ambiguous technical problems and driving them to resolution.
- Low ego and high empathy, with a history of growing engineers and supporting diverse, inclusive teams.
- Passion for AI safety and the role of security and risk management in building trustworthy AI systems.