Mercor is seeking AI Safety Experts fluent in English and Urdu to red team conversational AI models and identify vulnerabilities. This remote role involves generating human data by annotating failures, ensuring consistent testing, and documenting findings for actionable responses.

The ideal candidates will have prior experience in red teaming and the ability to explain risks clearly to diverse stakeholders. The role offers a compensation of $20–$22/hour and emphasizes a quick application process.