Software Security Engineer, Secure Frameworks

Anthropic•Seattle, WA

2d•$32,000 - $485,000•Hybrid

About The Position

Anthropic's mission is to create reliable, interpretable, and steerable AI systems that are safe and beneficial for users and society. The Secure Frameworks team is part of Anthropic's Security Engineering organization and is responsible for building shared security libraries and frameworks that the rest of the company utilizes. This includes owning critical security foundations such as workload identity and authorization, cryptographic frameworks, and the centralized access proxy for internal applications. The team's mission is to make secure behavior the default and prevent entire classes of vulnerabilities through careful design. They collaborate closely with various teams across Anthropic, own their shipped products, assist customers in adopting better security practices, and sometimes work embedded within research, infrastructure, or product teams. In this role, you will design, build, and operate security foundations used across Anthropic's fleet, from threat modeling through production, on-call, and adoption. Much of the work is greenfield due to rapid infrastructure growth and compressed timelines, requiring you to help define the architecture. Depending on your strengths, you will focus on one or two key areas while contributing to others.

Requirements

8+ years of software engineering experience building security-relevant systems in production, including leading complex initiatives independently
Strong programming skills in Go, Rust or Python
Deep understanding of authentication and authorization systems: OAuth 2.0 / OIDC, JWTs, service-to-service auth, policy-based access control
Hands-on experience with PKI, X.509, mTLS and workload identity
Working knowledge of applied cryptography: choosing and composing primitives correctly, key management, and the ways libraries fail in practice
Experience shipping and operating high-reliability services on Kubernetes across cloud and on-prem environments, including being on call for them
A track record of bringing clarity and ownership to ambiguous technical problems and driving them to resolution across teams
A strong sense of developer experience
Clear communication across all levels of the organization
Passion for AI safety and the role security engineering plays in building trustworthy AI systems

Nice To Haves

Built security frameworks or libraries adopted across an engineering organization, and done the work to drive that adoption
Built or operated a zero-trust / BeyondCorp-style access gateway or identity-aware proxy
Migrated a fleet to mTLS at scale, including certificate issuance and staged enforcement
Maintained or contributed to open-source cryptographic libraries
Designed authorization systems using policy languages or relationship-based models
Familiarity with ML infrastructure (training, inference serving)

Responsibilities

Build critical security foundations including cryptographic frameworks, mTLS infrastructure, secure serialization, and authorization systems, designed to prevent entire classes of vulnerabilities
Partner with product, research, infrastructure, and other security teams to ensure frameworks integrate smoothly with lower-layer security controls
Scope, design and build security services and libraries end-to-end, maintain them in production, and drive through ambiguous technical problems with minimal oversight
Build and operate the zero-trust access gateway that fronts Anthropic's internal services
Design and roll out authorization frameworks so services enforce least privilege consistently instead of each inventing its own access controls
Own cryptographic frameworks and libraries
Mentor engineers, contribute to hiring, and grow security engineering culture across Anthropic