French AI Data Trainer - Remote

About the role

We are looking for a detail‑oriented French AI Data Trainer to work remotely as an hourly contractor. You will evaluate AI‑generated responses, annotate safety‑critical content, and help improve the alignment of large language models.

Key responsibilities

• Assess AI outputs for accuracy, safety, and compliance with policy.
• Annotate and label datasets in French and English with high linguistic precision.
• Identify unsafe content such as hate, sexual material, violence, bias, misinformation, illegal activities, and malicious code.
• Compare multiple model responses and score them using safety and quality rubrics.
• Detect model failure patterns (evasion, escalation, unsafe normalization) and conduct red‑team/adversarial testing.
• Document findings, reasoning, and support the refinement of annotation guidelines.

Required profile

• Bachelor’s degree or equivalent experience in linguistics, communications, psychology, law, policy, security studies, or related fields.
• Native or near‑native proficiency in French (reading and writing) and minimum C1 English proficiency.
• Experience in trust & safety, content moderation, policy enforcement, or risk operations.
• Proven experience red‑team testing large language models and documenting safety boundaries.
• Strong analytical judgment, high attention to detail, and ability to work independently.

Required skills

• LLM red‑team testing
• Use of AI tools such as ChatGPT, Gemini, Perplexity
• Structured safety taxonomy labeling