AI Safety Red Teamer — English & Tamil Bilingual Specialist

About the role

We're building a global red team of language experts who help make AI systems safer by actively trying to break them. As an AI Safety Red Teamer fluent in both English and Tamil, you'll craft adversarial prompts, probe model responses for weaknesses, and help generate the high-quality safety data that shapes how AI models handle sensitive, ambiguous, or potentially harmful situations.

This is remote, flexible work for people who think critically about language, culture, and risk — and who want their expertise to directly influence how AI systems behave in the real world.

What you'll do

• Design adversarial and edge-case prompts in English and Tamil to stress-test AI model outputs
• Review and evaluate AI-generated responses for bias, misinformation, harmful content, and safety failures
• Document vulnerabilities with clear, structured feedback that engineering and safety teams can act on
• Apply cultural and linguistic nuance to identify risks that may not surface in English-only testing
• Collaborate with a distributed team of reviewers to maintain consistent quality and evaluation standards
• Iterate on prompt strategies based on model behavior and evolving safety guidelines

Requirements

• Native or near-native fluency in both English and Tamil (written and spoken)
• Strong analytical and critical thinking skills, with attention to nuance and edge cases
• Comfort reviewing sensitive or potentially disturbing text content as part of safety evaluation work
• Reliable internet access and ability to work independently on a remote, text-based workflow
• Excellent written communication skills for documenting findings clearly
• Prior experience with content moderation, linguistics, translation, QA, or AI evaluation is a plus

Compensation

This role pays $20–$22 per hour, with flexible remote scheduling. Compensation is based on experience and evaluation quality.