> Markdown version of [/jobs/ext/4622640-staff-application-security-engineer](https://www.wearedevelopers.com/jobs/ext/4622640-staff-application-security-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Staff+ Application Security Engineer - **Company:** Anthropic Limited - **Location:** San Francisco, CA, United States - **Experience:** Expert - **Salary:** $320,000.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Software System Penetration Testing, Static Program Analysis, Cyber Security, Python (Programming Language), Software Engineering, TypeScript, Software Vulnerability Management, Large Language Models, Claude Code, Multi-Agent Systems, Software Security, Containerization, Claude - **Published:** November 7, 2026 - **Apply:** https://www.juju.com/job/28_25304_b4cd6a72587fb0e4f25c6336a56be650 ## About the Role * Hands-on application and infrastructure security experience, including cloud and containerized environments * Production-quality coding ability in at least one of Python, Go, Rust, or TypeScript, with a track record of building durable systems rather than one-off scripts * Practical threat-modeling and vulnerability-identification skills - you've found and reasoned about real bugs in real systems, even if breaking isn't your primary mode * Demonstrated ability to operate with high autonomy and ambiguity - comfortable being handed a problem and a lot of latitude rather than a spec * Clear technical communication with both engineers and leadership, * 7+ years in application security, security engineering, or security-focused software engineering * Already use LLMs as a core part of how you work, with opinions about where they help and where they don't * Experience securing agentic, code-execution, or LLM-integrated systems specifically * Prior ownership of a bug bounty program, vulnerability disclosure program, or vulnerability-management infrastructure at scale * Background building security automation or developer-facing security tooling * Offensive security or penetration testing experience Representative projects, Minimum education: Bachelor's degree or an equivalent combination of education, training, and/or experience Required field of study: A field relevant to the role as demonstrated through coursework, training, or professional experience Minimum years of experience: Years of experience required will correlate with the internal job level requirements for the position Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices. ## Description Anthropic's Application Security team secures the systems that build, serve, and increasingly are Claude. The attack surface is unlike most AppSec work: multi-agent orchestration, sandboxed code execution, agents holding delegated credentials, untrusted tool output crossing trust boundaries - problems with little prior art and no off-the-shelf playbook. The way the team works is also different. We use Claude as our primary tool across every part of the job: it drives our static analysis, drafts and fixes vulnerabilities as pull requests, performs first-line bug bounty triage, and assists threat modeling for design reviews. The human work is the judgment layer - system-level reasoning, deciding what matters, and building the next thing the model can't do yet. This is a builder's role on a senior team. We hire engineers who ship production systems and clear a hands-on threat-modeling bar - people who can find the vulnerability but would rather build the system that finds them all. Every engineer owns a system end-to-end, and the team's work has shaped customer-facing product security - including Claude Code's security review tooling, its security guidance plugin, sandboxing, and auto mode., * Design, build, and operate Claude-powered security systems - LLM-driven code analysis, automated vulnerability remediation, AI-assisted threat modeling - and own one or more of them end-to-end, including the cross-functional relationships that come with it * Lead secure design reviews and threat modeling for novel AI systems, identifying risks that don't map to existing frameworks * Evolve a public bug bounty program where automation handles routine triage and root-cause work, and engineers handle escalations and corner cases * Partner with Product, Infrastructure, and Research teams as an embedded security owner - consulting on launches, shaping architecture, and influencing decisions where security is the constraint * Share an operational on-run rotation with the rest of the team - bounty escalations, incident response, and launch consults on systems serving Claude in production, * An increasingly autonomous vulnerability pipeline - LLM-driven code analysis finds the issue, scores it for real exploitability, and opens the fix PR, with humans as the review step rather than the author * Bug bounty operations where Claude handles first-line triage and drafting, and engineers focus on the reports that actually need judgment * AI-assisted threat modeling that generates intake questions, drafts the model, and recommends which design reviews need a human in the room * Automated dependency vulnerability remediation across Anthropic's codebase * The company-wide vulnerability dashboard and SLA enforcement layer every engineering team works against * Threat models and security architecture for agentic product surfaces - code execution sandboxing, agent identity and delegated auth, tool-use boundaries