AI Safety Expert – English, Odia

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

📋 Description

• Conduct red-team assessments on conversational AI models and agents by utilizing jailbreaks, prompt injections, misuse scenarios, bias exploitation, and multi-turn manipulations.

• Produce human data through the annotation of failures, classification of vulnerabilities, and identification of systemic risks.

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

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

• Evaluate AI outputs related to sensitive topics like bias, misinformation, or harmful behaviors.

• Identify vulnerabilities that automated testing methods may overlook.

• Enhance evaluation coverage and minimize unexpected issues during production.

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

⛳️ Requirements

• Proficient fluency in both English and Odia is essential.

• Strong judgment regarding language and content; capability to evaluate whether AI responses are accurate, comprehensive, and suitable, along with the ability to articulate the rationale.

• Keen ability to detect subtle errors, inconsistencies, and gaps.

• Consistent adherence to guidelines and quality standards.

• Capability to clearly communicate reasoning to both technical and non-technical audiences.

• Flexibility to adapt across various projects, task types, and client needs.

• Engagement as an independent contractor.

• Must not need H1-B or STEM OPT sponsorship.

• Preferred: experience in adversarial machine learning, including jailbreak datasets, prompt injection, RLHF/DPO attacks, or model extraction.

• Preferred: experience in cybersecurity, including penetration testing, exploit development, or reverse engineering.

• Preferred: experience in socio-technical risk, including harassment/disinformation probing, abuse analysis, or conversational AI testing.

• Preferred: background in psychology, acting, or writing to foster unconventional adversarial thinking.

🏝️ Benefits

• Fully remote position.

• Flexible working hours.

• Weekly payments via Stripe or Wise.

• Project durations may vary based on needs and performance.

• Access to wellness resources and clear guidelines for sensitive projects.

• Opportunity to gain experience in human data-driven AI red teaming.

• Direct involvement in enhancing the robustness, safety, and trustworthiness of AI systems.

• Collaboration with leading researchers in the field.

• Competitive compensation.

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