Remote AI Safety Red Team Specialist

Mercor; New York (NY); Remote; USD 90,000 - 130,000; Full time

Job summary

Mercor is assembling a remote AI safety red team to probe and stress-test conversational models using adversarial inputs. The role involves reviewing outputs on sensitive topics, with clear guidelines and wellness resources, and the work is entirely text-based.

You will annotate failures, classify vulnerabilities, and generate reproducible reports and datasets for customers to act on, following structured taxonomies and playbooks to ensure consistency.

Qualifications

• Native fluency in English and Malayalam required.
• Strong judgment about language, content, and context.
• Structured, guideline-driven approach to testing.

Responsibilities

• Red team conversational AI models and agents (jailbreaks, prompt injections).
• Generate high-quality human data by annotating failures and flagging risks.
• Follow taxonomies and playbooks to ensure consistent testing.
• Document reproducible reports, datasets, and attack cases.

Skills

English

Malayalam

Job description

Mercor is assembling a remote AI safety red team to probe and stress-test conversational models using adversarial inputs. The role involves reviewing outputs on sensitive topics, with clear guidelines and wellness resources, and the work is entirely text-based.

You will annotate failures, classify vulnerabilities, and generate reproducible reports and datasets for customers to act on, following structured taxonomies and playbooks to ensure consistency.