AI Safety Red Teamer — English & Assamese Bilingual Specialist

About the role

We're building a global red team of bilingual language experts to help make AI systems safer, more accurate, and less prone to harmful behavior. As an AI Safety Expert fluent in both English and Assamese, you'll work directly with AI model outputs — probing them with adversarial prompts, surfacing weaknesses, and generating the structured feedback data that helps train more robust, trustworthy AI systems.

This is sensitive, high-impact work. You'll regularly encounter content touching on bias, misinformation, and potentially harmful behaviors, and your judgment will directly shape how safely these models perform for real users across languages and cultures.

What you'll do

• Design and execute adversarial prompts ("red teaming") to expose vulnerabilities, biases, or failure modes in AI model outputs
• Evaluate AI-generated responses in both English and Assamese for safety, accuracy, tone, and cultural appropriateness
• Identify and document instances of harmful, biased, or misleading content generated by AI systems
• Write clear, structured feedback and annotations that can be used to retrain and improve model behavior
• Collaborate with a distributed team of reviewers and safety specialists to calibrate evaluation standards
• Flag edge cases and escalate high-risk findings according to established review protocols
• Maintain strict confidentiality and professionalism given the sensitive nature of the content reviewed

Requirements

• Native or near-native fluency in both English and Assamese (written and spoken)
• Strong analytical and critical thinking skills, with attention to nuance and context
• Comfort engaging with sensitive or challenging content in a professional, objective manner
• Excellent written communication skills for documenting findings and feedback
• Reliable internet access and the ability to work independently in a fully remote setting
• Prior experience in content moderation, linguistics, translation, journalism, or trust & safety is a plus
• Familiarity with AI/ML concepts or prior exposure to AI evaluation work is a plus but not required

Compensation

This role pays between $20–$22 per hour, based on experience and evaluation performance. Flexible, remote, part-time or full-time hours available.