AI Application Security Engineer

5 months ago

H1B sponsor

Responsibilities

Own secure development practices, including AuthN/AuthZ, secrets management, input handling, secure code review, CI/CD checks, and pre-deployment controls.
Conduct threat modeling and translate identified risks into concrete controls, architecture reviews, and security documentation.
Define application-layer security models for agents, including tool scoping, permission boundaries, trust boundaries, credential handling, response validation, and user-context protection.
Build reusable secure-by-default agent development patterns, review checklists, code-level guardrails, and automated security tooling.
Define data protection standards and safeguards for PHI, PII, government records, AI pipelines, agent outputs, access controls, output filtering, and audit logging.
Partner with product, ML, platform, and delivery teams to embed security into products shipped to regulated customer environments.
Use AI to accelerate threat modeling, security reviews, code analysis, and internal tooling.

5+ years of experience in application security or product security with hands-on production-system experience at scale.
Strong application security fundamentals, including OWASP Top 10, AuthN/AuthZ, secure SDLC, secrets management, secure integration patterns, and cryptography basics.
Understanding of agentic AI security at the product layer, including agent scoping, authorization, and risk review.
Experience protecting sensitive data through access controls, audit logging, and controls against exposure through integrations and AI-generated outputs.
Ability to build and ship security tooling and collaborate directly with engineering and delivery teams in regulated environments.
Bonus: experience with agent security, LLM application security, or authorization and guardrail systems for agentic pipelines.
Bonus: familiarity with FedRAMP, HIPAA, SOC 2, or ISO 27001.
Bonus: proficiency in Python, Go, or TypeScript for security tooling and automation.
Bonus: experience with SAST/DAST tooling or automated security checks in developer workflows at scale.

11-50 employees

We're building an AI platform and applications for the world's most important institutions. Learn more at https://brain.co/