Staff AI Security Engineer — Enterprise Guardrails

Affirm; St. Louis (MO); Remote; USD 204,000 - 264,000; Full time

Benefits offered by this job

Health coverage

Tech stipends

Flexible Time Off

ESPP

Job summary

Affirm is seeking an experienced security engineer to own the end-to-end security review of enterprise AI/LLM systems, evaluating architecture, data flows, and permissions while embedding security into design.

You will threat model, review prompts and MCP configurations, build secure test cases, and design guardrails with Python and IaC, collaborating with Legal, Privacy, Compliance, IT, and Engineering.

Qualifications

• Seasoned security professional with AI/LLM security design experience.
• Experience threat modeling AI/LLM apps and securing agentic systems.
• Proven ability to build governance artifacts and vendor risk assessments.

Responsibilities

• Lead and improve the enterprise AI security review process evaluating architecture, data flows, and design.
• Threat model AI/LLM systems and data flows to identify risks and drive remediation.
• Review source code, prompts, and tool manifests; help build secure test cases.
• Design and deploy guardrails and policy-as-code for AI security (Python, IaC).
• Evaluate third-party SaaS vendors’ AI capabilities and drive risk-based adoption.
• Identify AI security vulnerabilities and contribute to incident response playbooks.
• Lead cross-functional AI security initiatives and communicate with stakeholders.

Skills

AI security

Threat modeling

Security architecture

Python

IaC

Terraform

Kubernetes

AWS

OAuth2

SAML

Education

Bachelor's degree in CS/IS

Tools

Jira

Okta

CASB

Notion AI

Slack AI

Google Workspace

Terraform

Kubernetes

Job description

Affirm is seeking an experienced security engineer to own the end-to-end security review of enterprise AI/LLM systems, evaluating architecture, data flows, and permissions while embedding security into design.

You will threat model, review prompts and MCP configurations, build secure test cases, and design guardrails with Python and IaC, collaborating with Legal, Privacy, Compliance, IT, and Engineering.