Remote AI Safety & Adversarial Testing (Eng & Punjabi)

Mercor; Karachi Division; Remote; PKR 7,630,000 - 8,393,000; Full time

Job summary

Mercor is seeking AI Safety Experts to test and harden conversational AI systems. You will work remotely, applying red-team techniques to uncover jailbreaks, injection risks, and misuse scenarios, while producing actionable data and reports for customers.

Ideal candidates hold native fluency in English and Punjabi, with experience in AI adversarial work or cybersecurity, and the ability to explain risks to both technical and non-technical stakeholders.

Qualifications

• Native fluency in English and Punjabi.
• Experience in AI adversarial red teaming or cybersecurity.
• Ability to explain risks clearly to technical and non-technical stakeholders.
• Adaptability to work across multiple projects and customers.

Responsibilities

• Red team conversational AI models and agents focusing on jailbreaks, prompt injections, misuse cases, and bias exploitation.
• Generate high-quality human data, annotate failures, classify vulnerabilities, and flag 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.

Skills

English

Punjabi

Red teaming

AI adversarial work

Cybersecurity

Socio-technical probing

Job description

Mercor is seeking AI Safety Experts to test and harden conversational AI systems. You will work remotely, applying red-team techniques to uncover jailbreaks, injection risks, and misuse scenarios, while producing actionable data and reports for customers.

Ideal candidates hold native fluency in English and Punjabi, with experience in AI adversarial work or cybersecurity, and the ability to explain risks to both technical and non-technical stakeholders.