About the company
Our client is a cybersecurity company.
The role
Raydar is recruiting for this role on behalf of our client. Research offensive techniques against modern developer tools and automated assistants, identifying how they can be misused or compromised. Turn findings into detections, enforcement rules and published analysis that influence how security leaders think about emerging risk. Suited to a hands-on researcher with several years of security work who enjoys working independently on hard, frontier problems.
What you'll do
- Probe developer tools and automated assistants for weaknesses using adversarial methods such as prompt injection and misuse of connected tools.
- Run comparative studies of attack success across open-source and proprietary underlying models.
- Publish research and findings that are written for security leadership audiences.
- Convert findings into detections and enforcement policies, and investigate real incidents in customer environments.
- Track emerging frameworks and risks, and brief senior security stakeholders on what is being observed.
Requirements
What we're looking for
- Two to ten years of experience in adversarial testing, threat analysis, or building detections within a security practice.
- Background in security work performed at a security software vendor, inside a corporate security group, or on a research team at an AI developer.
- Hands-on experience attacking models or agent tooling through prompt injection, jailbreaks or tool misuse.
- Able to work on-site five days a week.
Bonus points
- Prior work at a security software vendor, within an internal corporate security function, or at a large technology peer.
- Published research, CVEs, conference talks or capture-the-flag results.
- Prior detected and reported vulnerabilities (CVEs).
Benefits
Compensation and benefits
- Base salary: $175K - $225K
- Equity
Location and work model
- San Francisco, CA, United States
- On-site, 5 days in-office per week
- Full-time