Staff Security Researcher, AI Security
Join a security research team developing methods to detect, assess, and mitigate risks from AI applications, agents, and services in enterprise environments. The role combines emerging-threat research with practical detections and defenses for an AI security platform. Hybrid work requires at least 3
Обязанности
- Lead research into AI security risks across enterprise applications, agents, and services
- Develop methods to detect and analyze AI tools, agents, and their behavior
- Design risk assessment frameworks for agent permissions, data access, tool use, and potential data leakage or abuse
- Research attacks such as prompt injection, tool abuse, excessive agency, and data exfiltration; propose mitigations and policy controls
- Work with engineering and product teams to turn research into AI security platform capabilities
- Track security and AI research and assess its relevance to the innovation pipeline
Требования
- At least 3 years of experience as a security researcher
- Strong understanding of AI and LLM architectures and integrations, including RAG, MCP, tool orchestration, and AI agents
- Understanding of operating system and networking internals
- Experience developing detection logic from security telemetry
- Familiarity with MITRE ATT&CK, MITRE ATLAS, OWASP Top 10 for LLM Applications, and threat modeling
- Programming and scripting proficiency in Python and at least one of C/C++, Go, or Rust
Будет плюсом
- Experience researching security products such as threat detection or data protection solutions
- macOS security research experience, including macOS internals
- Experience with agentic AI frameworks and their security risks
- Familiarity with AI security toolkits
- Security or AI security publications, conference talks, or CVEs
Условия и преимущества
- Career advancement and personal growth opportunities
- Training programs
- Performance-based rewards
- Hybrid schedule: 3 days in the office and 2 days working from home