Armenian AI Evaluator (Remote) — Flexible Freelance

Productive Playhouse; United States; Remote; USD 28,000 - 55,000; Part time

Job summary

Productive Playhouse is building a talent pool of Armenian speakers to test and evaluate leading AI chatbots. This freelance, project-based opportunity lets you choose tasks, set your own hours, and work with other clients as needed.

Open to freelancers outside the U.S., with work arriving in batches and clear task specifications in each SOW. You will interact directly with AI models to assess capabilities, safety, and helpfulness, and deliver structured evaluations and data that inform model

Qualifications

• Native or expert-level fluency in Armenian, as spoken in Armenia.
• Strong English reading, writing, and communication skills.
• 18 years of age or older.
• Hands-on experience using AI models or large language models.
• Comfortable with written and voice-based conversations and asking sharp follow-ups.
• Own smartphone, computer, reliable internet, and comfort with web-based tools.
• Able to pass language proficiency and technical literacy assessment.
• Responsive communicator with fast onboarding.

Responsibilities

• Test and evaluate AI chatbots or language models through structured conversations with goals.
• Provide feedback on capabilities, safety, and helpfulness.
• Submit deliverables (write-ups, ratings, screenshots, recordings) per task.
• Participate in iterative batch-based work with project-specific SOW.

Skills

Armenian language

English proficiency

AI model experience

Strong communication

Independent contractor mindset

Tools

AI evaluation tools

Job description

Productive Playhouse is building a talent pool of Armenian speakers to test and evaluate leading AI chatbots. This freelance, project-based opportunity lets you choose tasks, set your own hours, and work with other clients as needed.

Open to freelancers outside the U.S., with work arriving in batches and clear task specifications in each SOW. You will interact directly with AI models to assess capabilities, safety, and helpfulness, and deliver structured evaluations and data that inform model