AI Privilege Escalation Prompt Attack Researcher

About the role

Responsibilities

• Characterize privilege escalation attack surface for LLM systems with tool access and external integrations
• Develop novel attack methodologies: system prompt extraction, tool parameter injection, multi-turn manipulation
• Build evaluation benchmarks measuring LLM system resistance to privilege escalation attempts
• Investigate indirect prompt injection through retrieved context overriding operator instruction
• Develop architectural defense recommendations for enterprise LLM deployment security
• Publish research findings and open-source evaluation frameworks for community adoption

Requirements

• 6+ years in AI security research, red team operations, or adversarial ML with LLM security focus
• Deep expertise in prompt injection taxonomy, LLM tool use architectures, and agentic system design
• Strong Python and LLM API skills with adversarial attack implementation and evaluation
• Research track record with publications or significant contributions in AI security or adversarial ML
• Ability to work with enterprise clients translating research findings into actionable deployment security posture

Benefits

• Hybrid: 2 days NYC, 3 days remote
• Top-of-market compensation for AI security research expertise
• Comprehensive health benefits
• Significant research and conference budget
• Equity participation

Keep exploring related roles

Use these topic pages to discover more roles in the same hiring cluster.