ENGINEERING
AI Security Engineer
Secure the AI systems we build and the enterprises that run them. You will threat-model agentic architectures, harden LLM applications against prompt injection and data exfiltration, and give client security teams a playbook they trust.
- New York
- Engineering
- Full-time
What you’ll do
- Run security reviews and penetration tests against LLM applications and agent toolchains
- Design guardrail architectures: sandboxing, least-privilege tool access, output filtering, audit trails
- Respond to and investigate incidents involving AI systems in client environments
- Write secure-by-default patterns for our delivery teams and keep them current with the threat landscape
- Brief client CISOs on AI risk posture and remediation roadmaps
What we’re looking for
- 4+ years in information security, with hands-on exposure to LLM application security
- Working knowledge of the OWASP Top 10 for LLM applications and MITRE ATLAS
- Experience threat-modelling RAG and agent systems: tool permissions, data boundaries, injection paths
- Familiarity with cloud security across AWS, Azure and GCP AI services
- Understanding of compliance regimes clients bring to AI programmes (SOC 2, GDPR, HIPAA, EU AI Act)
What we offer
- Competitive compensation package
- Comprehensive health benefits
- Retirement savings plan
- Security certification reimbursement
- Flexible work arrangements