Ai Safety Specialist - Fully Remote | Upto $22/Hr

Mercor; Sydney; Remote; AUD 32.000 - 43.000; Teilzeit

Zusammenfassung

Mercor connects elite AI talent with leading labs and offers a contract role for AI Safety Experts—English & Gujarati. The position is remote and focuses on red-teaming conversational AI models, identifying jailbreaks, prompt injections, and bias exploitation.

You will generate high-quality data by annotating failures, applying taxonomies, and producing actionable reports to help customers improve AI model performance. Work independently to meet deadlines and maintain quality.

Qualifikationen

• Native fluency in English and Gujarati.
• Strong judgment about language accuracy and appropriateness.
• Rigorous attention to detail.
• Ability to follow guidelines and quality standards.
• Clear communication for technical and non-technical audiences.
• Adaptability across projects, task types, and customers.

Aufgaben

• Red team conversational AI models and agents by identifying jailbreaks, prompt injections, misuse cases, and bias exploitation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure by following taxonomies, benchmarks, and playbooks to maintain 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.

Kenntnisse

English & Gujarati

Jobbeschreibung

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 & Gujarati
Type: Contract
Compensation: $16–$22/hour
Location: Remote

Role Responsibilities

• Red team conversational AI models and agents by identifying jailbreaks, prompt injections, misuse cases, and bias exploitation.
• Generate high-quality human data by annotating failures, classifying vulnerabilities, and flagging systemic risks.
• Apply structure by following taxonomies, benchmarks, and playbooks to maintain 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

• Native fluency in English and Gujarati.
• Strong judgment about language and content accuracy, completeness, and appropriateness.
• Rigorous attention to detail, noticing subtle errors and inconsistencies.
• Ability to work to guidelines and quality standards consistently.
• Clear communication skills for technical and non-technical audiences.
• Adaptability across projects, task types, and customers.

Preferred

• Experience with Adversarial ML: jailbreak datasets, prompt injection, RLHF/DPO attacks, model extraction.
• Background in Cybersecurity: penetration testing, exploit development, reverse engineering.
• Understanding of socio-technical risk: harassment/disinfo probing, abuse analysis, conversational AI testing.
• Skills in creative probing: psychology, acting, writing for unconventional adversarial thinking.

Resources & Support

• For details about the interview process and platform information, please check: https://talent.docs.mercor.com/welcome
• For any help or support, reach out to: [e-mail]

PS: Our team reviews applications daily. Please complete your AI interview and application steps to be considered for this opportunity.