AI Safety Red Teamer — English & Telugu Bilingual Specialist

About the role

We're building a global red team of bilingual language experts to stress-test AI systems before they reach the real world. Working in English and Telugu, you'll act as an adversarial tester — probing large language models with challenging, edge-case, and sensitive prompts to uncover weaknesses in how they reason, respond, and behave. Your findings will directly shape safety guardrails, content moderation systems, and alignment training data used by AI teams around the world.

This is a remote, flexible-hours contract role ideal for detail-oriented individuals who are fluent in both English and Telugu and comfortable engaging thoughtfully with sensitive or controversial content in a structured, professional way.

What you'll do

• Design and submit adversarial prompts intended to surface bias, misinformation, harmful outputs, or unsafe model behavior
• Evaluate AI-generated responses in both English and Telugu for safety, accuracy, tone, and policy compliance
• Document vulnerabilities and edge cases with clear, structured written feedback
• Collaborate with a distributed team of reviewers and safety researchers to refine testing methodologies
• Apply detailed rubrics and guidelines consistently across large volumes of review work
• Flag emerging risk patterns and contribute to iterative improvements in red-teaming protocols
• Maintain strict confidentiality and professionalism when engaging with sensitive topics (bias, misinformation, harmful behavior scenarios)

Requirements

• Native or near-native fluency in both English and Telugu (written and spoken)
• Strong critical thinking skills and comfort engaging with sensitive, controversial, or adversarial content in a professional, objective manner
• Excellent written communication skills for documenting findings clearly
• Ability to follow detailed evaluation guidelines and apply them consistently
• Reliable access to a computer and stable internet connection for remote work
• High attention to detail and strong judgment when assessing nuanced content
• Prior experience in content moderation, linguistics, AI training data, translation, or trust & safety work is a plus but not required

Compensation

This role pays $20–$22 per hour, based on experience and performance, with flexible remote scheduling. Payment is issued for verified hours worked on assigned red-teaming and evaluation tasks.