Remote | Punjabi-English AI Safety Red Team Evaluator — $20–$30/hour

Related keywords: bilingual remote jobfully remote jobwriting remote job

Overview

The position is a part-time consulting opportunity at 24-MAG, aimed at Punjabi-English bilingual professionals skilled in AI safety evaluation. This role primarily focuses on red team testing, adversarial reviews, and offering structured feedback on sensitive, text-based AI outputs. It provides an engaging environment to contribute to AI safety, reliability, and overall improvement in systems affected by bias and misinformation.

Key Responsibilities

As a Punjabi-English AI Safety Red Team Evaluator, the successful candidates will participate in various tasks central to the role, including:

Bilingual AI Safety & Red Team Testing

• Review and stress-test English and Punjabi AI outputs for reliability and harmful-behavior risks.
• Evaluate conversational AI models across multi-turn dialogues to assess how they handle sensitive and edge-case prompts.
• Identify vulnerabilities that require enhanced safety controls or response quality improvements.

Vulnerability Classification & Risk Review

• Annotate failures and classify vulnerabilities while maintaining consistency through established benchmarks.
• Apply specified taxonomies to assess misuse cases and socio-technical risks.
• Generate high-quality human evaluation data for improvement efforts.

Reproducible Documentation & Evaluation Artifacts

• Create clear documentation, reports, and evaluation datasets to support model improvement.
• Ensure findings are documented reproducibly for review and further analysis.
• Communicate risks in a way that is comprehensible to both technical and non-technical stakeholders.

Ideal Candidate Profile

The ideal candidate must possess the following:

• Fluency in both English and Punjabi.
• Experience in AI red teaming or adversarial testing, with an understanding of cybersecurity principles.
• An analytical mindset that can adapt to needs across various safety categories and project types.
• Strong written communication skills are crucial to explain findings clearly and effectively.

Additionally, having a background in fields such as AI safety, linguistics, or policy can be beneficial for potential candidates. Experience with structured frameworks for testing is also advantageous.

Nice-to-Have Skills

While not required, the following qualifications can enhance a candidate's appeal:

• Exposure to adversarial machine learning concepts and different attack patterns can be valuable.
• Cybersecurity experience, including penetration testing or security assessments, which adds depth to a candidate's skill set.
• Involvement in areas concerning social risk analysis or behavioral safety in AI can further bolster a candidate's application.
• Familiarity with generating reproducible reports or datasets that aid structured risk assessments.

Work Environment and Compensation

This role is fully remote, allowing candidates flexibility in their schedule. Compensation will be competitive, ranging from $20 to $30 per hour, depending on expertise and project scope. The position requires a part-time commitment that may vary based on project availability, and it aims to leverage candidates’ bilingual skills and safety judgements.

Payments are arranged weekly through Stripe or Wise, ensuring a smooth financial transaction process for the evaluators. All work will be conducted in a text-based format that may include sensitive subjects, such as bias, misinformation, or harassment. Candidates will be informed about topic areas ahead of engagement, and participation in sensitive projects will consider the individual's comfort level.

About 24-MAG

24-MAG LLC is an organization that connects skilled professionals with remote consulting opportunities, particularly in fields that require technical evaluation and project-based workflows. By engaging with 24-MAG, professionals can gain substantial experience in the growing sector of AI safety evaluation while also contributing toward the development of safer, more reliable AI systems.

This can be an excellent opportunity for job seekers looking to leverage their language skills and technical expertise into meaningful work that impacts the future of AI technology. The role not only facilitates professional growth but also aligns with an individual's values in promoting responsible AI development.

This job offer was originally published on himalayas.app