AI Safety Red Teamer — English & Vietnamese Bilingual Specialist

About the role

We're building a global red team of bilingual AI safety experts to help make AI systems safer, more reliable, and more culturally aware. As a Red Teamer, you'll probe AI models with adversarial prompts, surface vulnerabilities, and generate high-quality safety data across both English and Vietnamese language contexts. Your work will directly shape how AI systems handle sensitive topics like bias, misinformation, and harmful content before they reach real-world users.

This is a fully remote, flexible opportunity ideal for detail-oriented linguists, researchers, or safety-minded professionals who want to have a tangible impact on responsible AI development.

What you'll do

• Design and execute adversarial prompts to test AI model behavior in English and Vietnamese
• Identify vulnerabilities related to bias, misinformation, toxicity, and unsafe outputs
• Evaluate and annotate AI-generated responses against safety and quality guidelines
• Document findings clearly to support model improvement and safety research
• Collaborate with cross-functional teams to refine red-teaming methodologies
• Provide cultural and linguistic context to ensure nuanced, accurate safety assessments in Vietnamese

Requirements

• Native or near-native fluency in both English and Vietnamese (written and spoken)
• Strong critical thinking and analytical skills, with attention to nuance and edge cases
• Comfortable engaging with sensitive or difficult content in a professional, objective manner
• Excellent written communication skills for documenting findings and feedback
• Reliable internet access and ability to work independently in a remote setting
• Prior experience in content moderation, linguistics, research, QA, or AI training data work is a plus

Compensation

Hourly rate of $17–$25, based on experience and evaluation performance. Flexible remote schedule with ongoing project-based engagement.