Senior Security Engineer, Product Security

Essential Job Duties and Responsibilities

• Adversarially test our AI and LLM-backed features. Design and run attacks against LLM-backed applications and agents — prompt injection, jailbreaks, tool abuse, data exfiltration — and turn findings into pass/fail criteria product teams will act on.
• Build and operate production security services. Backend services and internal tooling — APIs, streaming transports, proxy/CLI/chat interfaces — in whichever of TypeScript, Node.js, .NET, or Python fits the problem, held to the same bar as any other production service: test coverage, CI, dependency management. 
• Find new ways to automate the work. Notice when something we do by hand has become automatable, prototype it, and make the case— even when it means replacing a tool we bought last year. 
• Review pull request vulnerability findings. Triage what scanning and AI-assisted review surface across our stacks, separating real findings from noise. Go deep by hand on auth paths and high-risk changes, and feed what you learn back into the tooling. 
• Threat model from product designs. Review PRDs and technical designs before code exists, infer trust boundaries and data flows in unfamiliar domains, and raise security questions while the design is still cheap to change.
• Test by hand and validate what you find. Manual testing of web applications and APIs, triage for real exploitability, and retest fixes. Support the red team’s bug bounty and continuous penetration testing programs.  
• Keep the AppSec tooling estate running and low-friction. SAST/dependency scanning tuning, finding triage and routing, SSO and access management, and automating the repetitive parts so the program scales without headcount. 
• Secure the infrastructure your tooling runs on. IAM least-privilege scoping, secrets management, and container/network lifecycle — as infrastructure as code, with automated drift checks. 
• Enable engineers to do the right thing. Build security training and documentation engineers will actually use. 
• Evaluate tools and help set the AI bar. Run structured bake-offs of security products against defined requirements and help set the standards AI/agent systems must satisfy before reaching production. 
• Back up the rest of the security team. Support investigations, threat hunting, and incident response for the products you cover, and contribute to the vulnerability management lifecycle and security analytics platform.

Required Skills, Knowledge, and Abilities

You ship production code. Strong backend engineering in at least one modern language, with at least one service you built that others depend on — async patterns, HTTP APIs, and streaming transports are familiar ground. We work across TypeScript, Node.js, .NET, and Python; depth in one plus the willingness to move between them matters more than any particular stack on your résumé.
You can read code you didn’t write, across more than one language and stack, well enough to judge whether a reported finding is real, catch the ones tooling missed, and propose a fix the engineer can act on.
You know how identity and authorization actually fail: token exchange and scope handling, session lifetime and revocation, request signing, OAuth pitfalls, and network-layer issues like SSRF and DNS rebinding. We’re looking for reasoning that finds real bugs, not checklist recall.
You understand API standards and how to secure them: REST and GraphQL in practice, OpenAPI and schema contracts, input validation, rate limiting, gateway-level auth, and webhook and service-to-service verification.
Hands-on testing of web applications and APIs — manual, not just scanner-driven — plus the triage, the clear write-up, and the retest.
Threat modeling from written designs. You can read a PRD in an unfamiliar domain, infer trust boundaries and data flows, and ask the right questions while the answer is still cheap.
Working AWS and infrastructure-as-code competence: IAM scoping, secrets management, container/compute lifecycle, network egress control, and infrastructure defined as code.
Practical exposure to AI/LLM security. You have attacked an LLM-backed application or agent — at work, in a CTF, in published research, or in your own lab — and can tell us what you found and why it worked.
You write and speak for people who are not in security. Findings engineers act on, documentation they use, and explanations that hold up in front of a product manager, an executive, or Legal.
Preferred:
Having owned an AppSec tooling estate: SAST/SCA tuning, finding routing, false-positive reduction
Running structured vendor evaluations or proofs of concept
Contributing to security policy or standards, including for AI systems
Delivering security training or building hands-on learning environments
Depth in cryptography and key management
Detection engineering, incident response, or threat hunting exposure
An understanding of how SaaS products get built — roadmaps, prioritization, why the ship date exists. Prior product or engineering management experience is a plus, not an expectation.