Misuse Red Team - Research Engineer/Research Scientist
This role involves researching and developing novel attack vectors against large language model safeguards, including automated jailbreaking, data poisoning, and misuse of AI agents. The researcher will design and run experiments, build scalable evaluation tooling, and collaborate with frontier AI labs to identify and mitigate vulnerabilities. Work focuses on advancing the security of frontier AI systems through technical red teaming and publishing impactful findings.