Remote AI Safety Expert (English and Gujarati) – Contract
Mercor is seeking a contract AI safety expert fluent in English and Gujarati to red-team frontier AI models. The work involves testing vulnerabilities, evaluating failures, and creating reproducible datasets and attack cases.
विवरण
- Test conversational AI models and agents for jailbreaks, prompt injections, misuse scenarios, bias exploitation, and multi-turn manipulation
- Create human-generated data by annotating failures, categorizing vulnerabilities, and identifying systemic risks
- Use established taxonomies, benchmarks, and playbooks to conduct consistent evaluations
- Develop reproducible customer reports, datasets, and attack cases
- Assess AI outputs for sensitive issues including bias, misinformation, and harmful behavior
- Find weaknesses that automated testing does not detect
- Broaden evaluation coverage to help prevent unexpected production issues
- Contribute to projects that train and improve frontier AI systems
आवश्यकताएँ
- Native or fluent English and Gujarati language skills
- Sound judgment when evaluating language and content for accuracy, completeness, and appropriateness
- Ability to detect nuanced errors, inconsistencies, and omissions
- Consistent adherence to established guidelines and quality standards
- Clear communication of reasoning to both technical and non-technical audiences
- Flexibility across projects, task formats, and customer needs
- Engagement as an independent contractor
- H-1B and STEM OPT candidates are not eligible
- Preferred: adversarial machine learning experience with jailbreak datasets, prompt injection, RLHF/DPO attacks, or model extraction
- Preferred: cybersecurity experience in penetration testing, exploit development, or reverse engineering
- Preferred: socio-technical risk experience involving harassment or disinformation probing, abuse analysis, or conversational AI testing
- Preferred: psychology, acting, or writing experience that supports unconventional adversarial thinking
लाभ
- Work remotely from the United States
- Choose a flexible schedule that suits your availability
- Receive weekly payments through Stripe or Wise
- Opt in or out of higher-sensitivity projects
- Access clear guidance and wellness resources for sensitive-content assignments
- Competitive hourly compensation
- Reasonable accommodations available upon request
- Opportunity to earn up to $90 per successful referral, subject to applicable limits