AI Safety Red Teamer

Job overview

The AI Safety Red Teamer role is remote; the posting does not specify a narrower geographic scope. Compensation is $70–$84 per hour. Mercor connects technical and creative talent with AI research labs. The red teamer tests frontier AI models for vulnerabilities and unsafe behavior, documents findings, and contributes to safety benchmarks and reports. The work supports AI researchers in improving model alignment, robustness, and safety.

What you'll do

• Design adversarial prompts to stress-test AI models
• identify jailbreaks, unsafe behaviors, hallucinations, and policy failures
• evaluate robustness across sensitive domains
• document vulnerabilities
• contribute to safety benchmarking and red-teaming reports
• collaborate with AI researchers on model safety and robustness.

What we're looking for

Skills & competencies

ai safety red teamer

remote

mercor

ai safety

adversarial prompting

model evaluation

jailbreak testing

trust and safety

cybersecurity

ai alignment

$70–$84/hr

Work arrangement

• Weekend coverage required

Why this role

The posting describes this as a contract position and requires a bachelor’s degree or higher and five or more years of relevant professional experience.

About the employer

Mercor is hiring for this role. Industry: All Other Professional, Scientific, and Technical Services. Sector: 54.