The work
You will create self-contained coding tasks used to test AI models and review the models’ technical outputs. The work draws on software engineering research and practical research coding methods.
Assignments may include reproducible environments and Docker-based setups. You will also keep task scope, examples, and expected results clear and technically sound.
- Build research coding tasks in your area of software engineering.
- Review AI-generated code and research outputs for technical quality.
- Give technically grounded feedback on model responses.
- Work on test generation, program repair, static analysis, or mining software repositories.
- Use reproducible research methods when appropriate, including Docker-based environments.
What it pays and takes
This is remote, asynchronous contract work. It is open to candidates in the countries specified for the role and requires fluent English.
- Pay: $50 per hour.
- Work setting: Remote and asynchronous.
- Language: Fluent English.
- Education: A completed or in-progress PhD in computer science or a related field.
- Research record: Published research in software engineering or a comparable research area.
- Research output: Released research code through GitHub or another research artifact.
- Technical background: Practical depth in test generation, program repair, static analysis, or mining software repositories.
- Communication: Ability to explain technical judgments and evaluate model-generated code or research outputs.
- Helpful experience: First-author publications at ICSE, FSE, ASE, or ISSTA, plus Docker and reproducible research environments.
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 writing coding tasks and reviewing model responses. OpenTrain helps people find these roles, build a profile, and develop experience in AI training and data labeling.