AI Safety Expert - Adversarial ML

mercor; Malaysia; Remote; MYR 90,000 - 124,000; Full time

Job summary

Mercor is seeking AI Safety Experts to work remotely on English and Tamil language safety testing. You will red-team conversational AI models, annotate failures, and document vulnerabilities to help improve AI reliability for customers.

Candidates should demonstrate strong language judgment, attention to detail, and the ability to work independently in a remote, contractual role. Experience in adversarial ML and cybersecurity is preferred but not required.

Qualifications

• Native fluency in English and Tamil.
• Strong judgment about language accuracy, completeness and appropriateness.
• Rigorous attention to subtle errors, inconsistencies and gaps.
• Structured approach to work, adhering to guidelines and quality standards.
• Clear communication with both technical and non-technical audiences.
• Adaptability across projects, task types, and customers.

Responsibilities

• Red team conversational AI models and agents through jailbreaks, prompt injections, and misuse cases. Identify bias exploitation and multi-turn manipulation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure by following taxonomies, benchmarks, and playbooks to ensure consistent testing.
• Document reproducibly by producing reports, datasets, and attack cases that customers can act on.
• Work independently and asynchronously to meet deadlines while improving AI model performance.

Skills

English

Tamil

Language judgment

Attention to detail

Structured work

Clear communication

Adaptability

Job description

About the job Mercor connects elite creative and technical talent with leading AI research labs. Headquartered in San Francisco, our investors include Benchmark , General Catalyst , Peter Thiel , Adam D'Angelo , Larry Summers , and Jack Dorsey.

Position: AI Safety Experts - English & Tamil

Type: Contract Compensation: $16-$22/hour

Location: Remote

Responsibilities

• Red team conversational AI models and agents through jailbreaks, prompt injections, and misuse cases. Identify bias exploitation and multi-turn manipulation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure by following taxonomies, benchmarks, and playbooks to ensure consistent testing.
• Document reproducibly by producing reports, datasets, and attack cases that customers can act on.
• Work independently and asynchronously to meet deadlines while improving AI model performance.

Qualifications

• Must-Have
  • Fluent Language Skills Required: English & Tamil. Native fluency in English and Tamil is required.
  • Strong judgment about language and content accuracy, completeness, and appropriateness.
  • Rigorous attention to subtle errors, inconsistencies, and gaps.
  • Structured approach to work, adhering to guidelines and quality standards.
  • Clear communication with both technical and non-technical audiences.
  • Adaptability across projects, task types, and customers.
• Preferred
  • Experience in Adversarial ML : jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction.
  • Cybersecurity skills: penetration testing, exploit development, reverse engineering.
  • Knowledge of socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing.
  • Creative probing skills: psychology, acting, writing for unconventional adversarial thinking.