Amazon
Security Engineer, Security for Employee and Legal Systems (SEALS)
$159K–$202K
Seattle, WA · Full-time · Posted Sep 28, 2026
- Experience
- 3+ years programming; 2+ years in each of the specified security-code-review and systems troubleshooting areas. STEM degree or IT Security experience is accepted.
- Education
- Bachelor's degree in a STEM field (Science, Technology, Engineering, Mathematics), or experience in IT Security.
About the job
Secures Amazon’s HR and Legal applications by partnering with builder teams, reducing risks throughout development, and addressing security issues in GenAI and agentic workflows.
What you'll do
- Threat-model GenAI applications and secure agentic workflows, including authentication and authorization.
- Review designs and code, guide remediation, and create automated detections that provide security feedback during development.
- Own security outcomes for assigned teams, promote secure development paths, and communicate risks and outcomes to stakeholders.
What they ask for
- 3+ years programming in a listed language or similar object-oriented language.
- 2+ years of scripting, programming, and security code review in a common programming language (non-internship).
- 2+ years troubleshooting systems, analyzing logs, or automating tasks with command-line tools (non-internship).
- Bachelor’s degree in a STEM field, or experience in IT Security; knowledge of security vulnerabilities and remediation techniques.
Nice to have
- 2+ years in a combination of threat modeling, secure coding, identity management and authentication, software development, cryptography, system administration, or network security.
- Knowledge of HTTP(S), DNS, and TCP/IP; AWS experience; or security activities across the SDLC.
Pay
159,300.00 - 202,400.00 USD annually (base salary)
Tools
Python, Ruby, Go, Swift, Java, .Net, C++, AWS
AI focus
Threat-model and secure GenAI/LLM applications and agentic workflows; use GenAI for code review, detection authoring, and triage.
Model-written summary · the original posting is authoritative