AI Safety Red Team Engineer - Remote

Obsidian; New York (NY); Remote; USD 85,000 - 120,000; Full time

Benefits offered by this job

Experience in AI red teaming

Role in enhancing AI robustness

Job summary

Obsidian is looking for experts in red teaming to help enhance AI safety through adversarial probing. The role involves identifying vulnerabilities in conversational models and generating human data to strengthen customer AI systems.

Applicants should have experience in adversarial work in AI and cybersecurity, alongside strong communication skills. The position is remote, allowing you to make a significant impact in AI safety and robustness.

Qualifications

• Prior red teaming experience (AI adversarial work, cybersecurity, socio-technical probing).
• Curiosity and adversarial mindset to push systems to breaking points.
• Ability to clearly explain risks to technical and non-technical stakeholders.

Responsibilities

• Red team conversational AI models and agents through various testing methodologies.
• Generate and annotate high-quality human data; document findings.
• Follow benchmarks and frameworks to maintain consistent testing.

Skills

Red teaming experience

Adaptability

Clear communication

Job description

Obsidian is looking for experts in red teaming to help enhance AI safety through adversarial probing. The role involves identifying vulnerabilities in conversational models and generating human data to strengthen customer AI systems.

Applicants should have experience in adversarial work in AI and cybersecurity, alongside strong communication skills. The position is remote, allowing you to make a significant impact in AI safety and robustness.