Security Research Engineer - Project Perception
Core
Build realistic attack chains and automated graders to test AI systems' ability to distinguish threats from benign activity in cloud, on-premises, and hybrid environments.
Role type
Senior Security Research Engineer (AI Safety & Adversarial Testing)
Builds
Verifiable attack scenarios, automated graders, and reward signals for AI training and evaluation
Domain
Cybersecurity, AI Safety, Cloud Security
Deliverable
production ML models
Required skills
Vulnerability research, authorized offensive security testing, security tool development, multi-step attack chain construction, cloud IAM assessment, source-code analysis, reverse engineering, telemetry analysis, scenario automation
Preferred skills
Experience with red-team/purple-team exercises, identity system reproduction, fix verification, benign lookalike creation
Technologies
Python, C#, Go, C/C++, TypeScript, Azure, AWS, GCP, Windows, Linux
Responsibilities
Reproduce real vulnerabilities and adversary techniques; build multi-step attack chains with verified prerequisites; create automated graders and reward signals; assess cloud IAM and test authentication/permissions; confirm security findings via source-code review and controlled tests; run authorized security tests with isolation; reconstruct attacks from incident evidence; create attack variants and patched controls to test detection accuracy.
Seniority
Senior, hands-on IC
