SWE Bench Engineer (Golang)

You will help evaluate how large language models handle realistic bug-fixing tasks in public open-source repositories. The work combines hands-on Golang development, repository setup, issue analysis, and software-quality judgment.

• Analyze and triage issues across open-source libraries.
• Set up code repositories and development environments with Docker.
• Review unit-test coverage and assess test quality.
• Run and modify codebases locally to evaluate bug-fixing performance.
• Work with researchers to identify repositories and issues that are difficult for large language models.

What it pays and takes

This is a fully remote contractor assignment with part-time employment status. The assignment is expected to last 2 to 3 weeks.

• $15 per approved task.
• At least 20 hours per week.
• Open to candidates in Argentina, Bangladesh, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, Ecuador, Egypt, El Salvador, Ghana, Guatemala, Guyana, Haiti, Honduras, India, Mexico, Nicaragua, Nigeria, Pakistan, Panama, Paraguay, Peru, Puerto Rico, Suriname, Turkey, Uruguay, and Venezuela.
• Fluent in English.
• Experience level: entry level.
• At least three years of software engineering experience and strong Golang skills.
• Proficiency with Git, Docker, and basic software pipeline setup.
• Ability to navigate complex codebases and run, modify, and test real-world projects locally.
• Familiarity with public GitHub repositories, open-source contribution or evaluation, and software-quality assessment is valuable.
• Experience with LLM research, developer tools, or automation agents is helpful.

How it works

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

About AI training work

AI training is the human work behind systems that learn from examples, including code tasks and evaluations of model responses. Engineers are needed for specialist work because they can judge whether code, tests, repositories, and proposed fixes are technically sound.