AI Safety Experts – English, Marathi

This is a fully remote position, open to applicants in United States.

📋 Description

• Conduct red-team assessments on conversational AI models and agents utilizing jailbreak techniques, prompt injections, misuse scenarios, bias exploitation, and multi-turn manipulation.

• Create human-generated data by annotating failures, categorizing vulnerabilities, and identifying systemic risks.

• Implement taxonomies, benchmarks, and playbooks to ensure consistent testing methodologies.

• Develop reproducible reports, datasets, and attack scenarios for clients.

• Review AI outputs related to sensitive issues such as bias, misinformation, and harmful behaviors.

• Identify vulnerabilities that automated testing may overlook.

• Broaden evaluation coverage and minimize unexpected outcomes in production.

• Assist Mercor clients in enhancing the safety, robustness, and trustworthiness of their AI systems.

⛳️ Requirements

• Must possess native fluency in both English and Marathi.

• Strong judgment regarding language use and content; capable of evaluating the accuracy, completeness, and appropriateness of AI responses.

• Proficient in spotting subtle errors, inconsistencies, and gaps in information.

• Consistently able to adhere to guidelines and quality standards.

• Capable of articulating reasoning effectively to both technical and non-technical audiences.

• Flexible in adapting to various projects, task types, and client needs.

• Must hold independent contractor status.

• Should not require H1-B or STEM OPT support.

• Nice-to-have: experience in adversarial machine learning, including working with jailbreak datasets, prompt injections, RLHF/DPO attacks, or model extraction.

• Nice-to-have: background in cybersecurity, including penetration testing, exploit development, or reverse engineering.

• Nice-to-have: experience in socio-technical risks, such as harassment/disinformation probing, abuse analysis, or testing of conversational AI.

• Nice-to-have: experience in psychology, acting, or writing for unconventional adversarial thinking.

🏝️ Benefits

• Fully remote position that allows for flexible scheduling.

• Weekly compensation through Stripe or Wise.

• Optional participation in higher-sensitivity projects.

• Clear guidelines and wellness resources available for handling sensitive content.

• Competitive remuneration.

• Opportunity to collaborate with leading researchers in the field.

• Referral bonuses of up to $90 for each successful referral (conditions apply).