Responsibilities

• Craft realistic, chat-style developer prompts covering code review, debugging, error diagnosis, configuration, and related scenarios.
• Source and adapt real pull requests and engineering discussions to create authentic evaluation tasks.
• Write clear, technically accurate model responses that demonstrate strong reasoning and explanation quality.
• Evaluate AI outputs based on logic, clarity, and technical judgment rather than executable correctness.
• Apply software engineering expertise to ensure prompts and responses reflect real-world developer workflows.
• Contribute to structured evaluation datasets used to assess AI reasoning and communication quality.
• Collaborate asynchronously with research teams to refine task quality and consistency.

Qualifications

• Strong experience in software engineering, technical research, or educational content development.