French AI Conversation Evaluator — Remote, Flexible Hours

Outlier; Strasbourg; Remote; EUR 25,000 - 36,000; Part time

Benefits offered by this job

Weekly pay via PayPal or Airtm

Flexible schedule

Remote work from home

Model Playground access

Surge pricing incentives

Job summary

Outlier is a platform that connects subject-matter experts with paid AI training projects conducted remotely. As a French-language expert, you will participate in live scenarios with an AI voice model, assess conversations, and provide detailed ratings to improve AI performance.

Experts control their schedules, typically contributing 5–10 hours per week, with potential to scale to 40 hours. Weekly pay is supported, and additional incentives may apply.

Qualifications

• Native or highly fluent French speaker.
• Experience with improvisation, role-play, or unscripted performance (acting, improv, podcasting, voice work).
• Customer-facing experience where you regularly handle unscripted conversations (customer service, sales, coaching, interviewing).
• Experience evaluating or reviewing conversations, calls, or audio content (QA, call auditing, content moderation).

Responsibilities

• Participating in a live, scenario-based conversation in French with an AI voice model, playing out a natural, realistic exchange.
• Listening to and reviewing recorded French-language conversations between a speaker and an AI model, checking whether the model follows the scenario and any relevant guidelines.
• Scoring conversations across dimensions like naturalness, helpfulness, and audio quality, then explaining your ratings in detail.
• Choose the right response from paired options.
• Write prompts to initiate AI model conversations.

Skills

French fluency

Improv/Role-play

Customer-facing experience

QA / content review

Quality evaluation

Job description

Outlier is a platform that connects subject-matter experts with paid AI training projects conducted remotely. As a French-language expert, you will participate in live scenarios with an AI voice model, assess conversations, and provide detailed ratings to improve AI performance.

Experts control their schedules, typically contributing 5–10 hours per week, with potential to scale to 40 hours. Weekly pay is supported, and additional incentives may apply.