Staff AI Security Engineer (remote)

FiveSix Partners is a well-funded, early-stage AI startup focused on enabling organizations to deploy trustworthy and secure AI systems. They are seeking a staff-level AI Security Engineer to own the company's security function, research emerging attack techniques, and develop security capabilities for AI applications.

Responsibilities:

• Design and build security capabilities for AI-native applications and autonomous AI agents
• Research emerging AI threats, including prompt injection, tool poisoning, jailbreaks, indirect prompt injection, model manipulation, memory poisoning, sandbox escapes, and agent privilege escalation
• Develop AI security tooling, evaluation frameworks, and automation to identify vulnerabilities before they reach production
• Perform adversarial assessments and red team exercises against LLM-powered applications and agentic workflows
• Lead the company's application security program, including secure design reviews, threat modelling, secure coding practices, code reviews, and vulnerability management
• Partner closely with engineering to embed AI security and application security throughout the software development lifecycle
• Help define the long-term security strategy and technical roadmap as the company grows

Requirements:

• AI Security, LLM Security, or Agentic AI Security
• Building or securing AI applications powered by modern foundation models
• Adversarial testing, AI red teaming, prompt injection research, or AI vulnerability research
• Application security and secure software development practices
• Security engineering or software engineering with strong programming skills (Python, Go, Rust, or similar)
• Cloud-native infrastructure, Kubernetes, containers, and modern application security