Software Engineering AI Task Author

The Work

You will create and evaluate realistic software engineering tasks that help AI systems learn from clear human expertise. Your work will cover codebases, logs, alerts, integrations, configuration issues, and engineering task environments.

You will also solve each task, implement and verify the fix, adjust difficulty, and review AI-drafted prompts and rubrics for accuracy and consistency.

• Write task prompts about integration failures, log-based debugging, source-code bug traces, configuration problems, and integration fixes.
• Create scoring rubrics that define the correct diagnosis and fix, including how the result should be checked.
• Set up realistic environments with codebases, logs, alerts, and system states.
• Implement and verify relevant fixes in live or simulated environments.
• Adjust task complexity and ambiguity to calibrate difficulty.
• Review AI-generated prompts and rubrics for accuracy and consistency.

What It Pays and Takes

This is a remote, hourly contract role with task-based scheduling. The role is listed as entry level, but the work requires hands-on, industry-level experience as a working software engineer.

• Pay: $70 to $120 per hour
• Hours: 20 or more hours per week
• Location: United States only
• Language: English
• Work type: Part-time contract
• Required: Strong software engineering and debugging experience
• Required: Git, version control workflows, and code review experience
• Required: Experience reading logs, using observability tools, and tracing failures through codebases
• Required: Experience with infrastructure, integrations, APIs, services, configuration files, and CI/CD systems
• Required: Ability to implement and verify fixes and define precise, checkable correctness criteria
• Required: Careful attention to detail and clear communication of technical reasoning
• Helpful background: Enterprise SaaS, financial technology, distributed systems, monitoring platforms, observability tools, or developer-focused AI tools

How It Works

Apply on OpenTrain with your resume and then complete the application on the hiring site.

About AI Training Work

AI training work is the human side of building artificial intelligence. People create examples, write and rate responses, review code, and set clear standards so AI systems can learn what good work looks like. Experienced specialists are paid for the technical judgment needed to make those examples accurate and useful.