AI Safety Red Teamer — English & Punjabi Bilingual Specialist
Paid remote expert bounty in Expert network (Global). Work in the software you already know; rates and ladder published up front.
- $30-$250
- per hr, published up front
- 1
- slots left of 1
- Remote
- no AI experience required
What you'll be paid to show
About the role
We're building a global red team of bilingual language experts to help make AI systems safer, more reliable, and less prone to harmful or biased outputs. As an AI Safety Expert fluent in both English and Punjabi, you'll work directly with AI-generated content, probing models with adversarial prompts, evaluating responses for risk, and helping generate the high-quality safety data that trains the next generation of AI systems.
This is remote, flexible, freelance work suited to detail-oriented thinkers who care about the responsible development of AI and want to apply their language skills to meaningful, high-impact work.
What you'll do
- Craft adversarial prompts designed to surface model vulnerabilities, biases, and unsafe behaviors
- Review and annotate AI-generated outputs in both English and Punjabi for safety, accuracy, and appropriateness
- Identify instances of misinformation, harmful content, bias, or policy violations in model responses
- Document findings clearly and consistently to support model improvement pipelines
- Collaborate with a distributed team of reviewers and safety specialists on evolving guidelines and edge cases
- Provide linguistic and cultural context to help calibrate model behavior for Punjabi-speaking audiences
Requirements
- Native or near-native fluency in both English and Punjabi (written and spoken)
- Strong critical thinking and ability to evaluate nuanced, sensitive content objectively
- Comfortable engaging with content related to bias, misinformation, and potentially harmful topics in a professional capacity
- Excellent attention to detail and consistency in written feedback
- Reliable internet access and ability to work independently in a remote setting
- Prior experience in content moderation, translation, linguistics, data annotation, or AI training data work is a plus
Compensation
This role is compensated hourly at competitive market rates, with flexible engagement based on project needs and availability.
What we're looking for
- AI safety evaluation
- Bilingual content review (English/Punjabi)
- Red teaming / adversarial testing
- Data annotation
- Bias and misinformation detection
- Content moderation
- Written communication