AI Safety LLM Evaluator, French and English

The work

You will review AI-generated responses and create safety-focused evaluation content in English and French. Your work will help set clear standards for identifying toxic, unsafe, or policy-violating model behavior.

• Score and annotate AI-generated text.
• Evaluate responses for reasoning, safety, and policy alignment.
• Create and curate red-team training cases, including nuanced or potentially explicit content.
• Help improve labeling standards for AI safety evaluation.

What it pays and takes

This is a fully remote, hourly contractor role for someone with intermediate experience and LLM red-team experience. The work requires strong English and French skills and careful judgment when reviewing sensitive content.

• Pay: $24 to $36 per hour, with a listed rate of $30 per hour.
• Hours: 20 or more per week.
• Location: Worldwide and fully remote.
• Work type: Part-time contractor.
• Languages: English and French.
• Experience: Intermediate level with LLM red-team experience required.

How it works

Apply on OpenTrain, and the employer reviews applications there.

About AI training work

AI training work is the human side of building artificial intelligence. People review and label examples so models learn to produce safer, more useful responses, and experienced evaluators are paid for applying careful judgment to difficult cases.