General Motors
Senior Security Software Engineer, Advanced Security
Warren, Michigan, United States of America +1 · Hybrid · Senior · Full-time · Posted Oct 28, 2026
- Location
- Warren, Michigan, United States of America
Austin, Texas, United States of America - Experience
- 5+ years in software engineering, cloud engineering, or security engineering.
- Education
- Bachelor’s degree in Computer Science, Cybersecurity, Engineering, or a related field, or equivalent practical experience.
- Visa sponsorship
- GM does not provide immigration-related sponsorship for this role; applicants must not need GM immigration sponsorship now or in the future.
About the job
Build and operate secure cloud services and AI agents that analyze security findings, test authorized targets, and support remediation across GM.
What you'll do
- Design and operate production cloud services, infrastructure, and AI agents for security analysis.
- Build sandboxed testing environments and control agent access to tools, data, and computing resources.
- Evaluate AI model performance and collaborate across security teams to deliver improvements, communicate risks, and mentor engineers.
What they ask for
- Bachelor’s degree in Computer Science, Cybersecurity, Engineering, or a related field, or equivalent practical experience.
- Strong experience writing production Python.
- Experience with infrastructure as code and CI/CD on a major cloud platform.
- Working knowledge of application security, secure systems practices, and agentic LLM tools and solutions.
Nice to have
- Experience with LLM applications and agent frameworks, containers and Kubernetes, sandboxing, or static/program analysis.
- Experience in vulnerability research, penetration testing, or exploit development; familiarity with AI security risks and controls.
Tools
Python, Terraform, CI/CD, Kubernetes, Semgrep, CodeQL, OWASP Top 10 for LLM Applications
AI focus
Build AI agents to triage security findings, gather evidence, and propose fixes; evaluate model accuracy, cost, and latency; secure agent access and defend against prompt injection.
Model-written summary · the original posting is authoritative