Remote AI Safety Red Team Specialist

Mercor; Karachi Division; Remote; PKR 22,191,000 - 33,287,000; Full time

Job summary

Mercor is building a red team for a project that probes AI models with adversarial inputs. The work involves reviewing AI outputs touching sensitive topics such as bias, misinformation, or harmful behaviors.

All work is text-based, and participation in higher-sensitivity projects is optional and guided by clear guidelines and wellness resources. In this role you will red-team conversational AI models, generate high-quality human data, follow taxonomies and playbooks to ensure testing

Qualifications

• You bring strong judgment about language and content: you can tell whether an AI response is accurate, complete, and appropriate, and explain why
• You’re rigorous: you notice subtle errors, inconsistencies, and gaps that others skim past
• You’re structured: you work to guidelines and quality standards consistently, not ad hoc
• You’re communicative: you explain your reasoning clearly to technical and non-technical audiences
• You’re adaptable: you thrive moving across projects, task types, and customers

Responsibilities

• Red team conversational AI models and agents: jailbreaks, prompt injections, misuse cases, bias exploitation, multi-turn manipulation
• Generate high-quality human data: annotate failures, classify vulnerabilities, and flag systemic risks
• Apply structure: follow taxonomies, benchmarks, and playbooks to keep testing consistent
• Document reproducibly: produce reports, datasets, and attack cases customers can act on

Job description

Mercor is building a red team for a project that probes AI models with adversarial inputs. The work involves reviewing AI outputs touching sensitive topics such as bias, misinformation, or harmful behaviors.

All work is text-based, and participation in higher-sensitivity projects is optional and guided by clear guidelines and wellness resources. In this role you will red-team conversational AI models, generate high-quality human data, follow taxonomies and playbooks to ensure testing