Remote Software Trainer (AI Code Evaluation)

About the role

We are seeking software engineers across Latin America to join a fully remote AI training program. As a Software Trainer you will evaluate AI‑generated code, rate the quality of AI responses and create coding challenges that help improve the model’s performance.

Key responsibilities

• Review and assess code produced by AI models for correctness, efficiency and best‑practice compliance.
• Rate AI responses on dimensions such as accuracy, instruction‑following and overall quality.
• Design original computer‑science questions and coding challenges to test AI capabilities.
• Identify bugs, errors and inconsistencies in AI‑generated code using structured guidelines.
• Provide clear, constructive written feedback to improve future AI outputs.
• Follow project guidelines, quality standards and confidentiality requirements.

Required profile

• Software engineering experience at junior, mid‑level or senior level.
• Proficiency in one or more programming languages (Python, Java, JavaScript, TypeScript, C++, Swift, Rust).
• Strong attention to detail, analytical thinking and problem‑solving abilities.
• Background in full‑stack, backend, frontend, mobile or any engineering discipline.
• Legal authorization to work in any of the listed Latin American countries.

Required skills

• Python
• Java
• JavaScript
• TypeScript
• C++
• Swift
• Rust

What we offer

• Compensation up to $34 USD per hour, with performance incentives averaging 7.5 %.
• Weekly payments via PayPal or AirTM.
• Fully remote, asynchronous work with flexible workload based on project availability.
• Free access to leading AI tools valued at over $200 per month.