AI Safety Specialist - Red Team & Risk Analysis (Remote)

eBSI Export Academy; EE. UU.; A distancia; USD 23.000 - 34.000; A tiempo parcial

Descripción de la vacante

Mercor is seeking AI Safety Experts fluent in English and Vietnamese for a remote contract role. You will red team conversational AI models, craft jailbreaks, annotate failures, and build reproducible reports and datasets for customer action.

Applicants should have experience in red teaming, AI adversarial work, or cybersecurity, and be able to explain risks clearly to both technical and non-technical stakeholders. Preference for adversarial ML and socio-technical risk knowledge.

Formación

• Fluent in English and Vietnamese.
• Prior experience in red teaming, AI adversarial work, or cybersecurity.
• Ability to explain risks clearly to both technical and non-technical stakeholders.

Responsabilidades

• Red team conversational AI models and agents by conducting jailbreaks, prompt injections, and bias exploitation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure using taxonomies, benchmarks, and playbooks to ensure consistent testing.
• Document reproducibly by producing reports, datasets, and attack cases for customer action.
• Review AI outputs on sensitive topics like bias and misinformation, with optional participation in higher-sensitivity projects.

Conocimientos

Fluent in English and Vietnamese

Red teaming / AI adversarial work

Explain risks clearly to all audiences

Descripción del empleo

Mercor is seeking AI Safety Experts fluent in English and Vietnamese for a remote contract role. You will red team conversational AI models, craft jailbreaks, annotate failures, and build reproducible reports and datasets for customer action.

Applicants should have experience in red teaming, AI adversarial work, or cybersecurity, and be able to explain risks clearly to both technical and non-technical stakeholders. Preference for adversarial ML and socio-technical risk knowledge.