Generative AI Red Team Engineer

JPMorgan Chase & Co.; Plano (TX); On-site; USD 61,000 - 101,000; Full time

Job summary

JPMorgan Chase & Co. is seeking a Security Engineer within Cybersecurity & Technology Controls for AI/ML to design secure AI architectures and strengthen red teaming programs.

You will work across product, data science, cyber, legal, and risk teams to assess AI use cases, build internal tools, dashboards, and automation for red teaming efforts. You will apply threat modeling, penetration testing, and adversarial testing practices to complex AI systems, including RAG pipelines and ML workloads,

Qualifications

• We seek formal training or certification in public cloud concepts and hands-on cloud experience.
• Experience with threat modeling, vulnerability testing, and adversarial testing for AI/ML systems.
• Proficiency in Python scripting and building security tooling for red teaming.
• Familiarity with MITRE ATLAS, OWASP Top 10 for LLMs, IAM, OIDC, and SAML is preferred.

Responsibilities

• We develop and strengthen security strategies, red teaming programs, and scalable solutions.
• We design secure AI and software architectures, review designs and code for adversarial resilience.
• We reduce AI/LLM security risks by applying standards and evolving controls.
• Partner with product, data science, cyber, legal, and risk teams to assess AI use cases and risk.

Skills

Python scripting

Cloud concepts

Threat modeling

Adversarial testing

AI/ML concepts

Education

Formal training or certification in public cloud concepts

Tools

Terraform

CloudFormation

PyRIT

Garak

LLM evaluation harnesses

Job description

JPMorgan Chase & Co. is seeking a Security Engineer within Cybersecurity & Technology Controls for AI/ML to design secure AI architectures and strengthen red teaming programs.

You will work across product, data science, cyber, legal, and risk teams to assess AI use cases, build internal tools, dashboards, and automation for red teaming efforts. You will apply threat modeling, penetration testing, and adversarial testing practices to complex AI systems, including RAG pipelines and ML workloads,