The work
You will create, solve, review, and validate challenging software-engineering tasks used to train AI systems. The work covers front-end and back-end systems, unfamiliar codebases, debugging, code translation, testing, and review of AI-generated patches.
- Explore unfamiliar software repositories and trace data and execution across interfaces, services, libraries, APIs, databases, and external dependencies.
- Build and debug front-end and back-end components within existing architectures and contribution standards.
- Reproduce software issues, find their underlying causes, and implement suitable solutions.
- Translate complete coding builds from one programming language into another.
- Write unit, integration, regression, and end-to-end tests.
- Review AI-generated patches for correctness, completeness, security, and maintainability.
What it pays and takes
This is a fully remote contractor role with flexible scheduling. The expected commitment is approximately 15 hours per week, with the option to work weekends when needed.
- $100 to $150 per hour equivalent, paid per task that meets the project specifications.
- Approximately 15 hours per week.
- Fully remote, with applications open to candidates in the eligible countries shown for the role.
- Professional fluency in English.
- A public or otherwise verifiable GitHub profile.
- Meaningful ownership of or contributions to one or more open-source repositories.
- Strong software-engineering experience in front-end, back-end, full-stack, systems, developer tooling, infrastructure, or another substantial technical area.
- Experience with multiple programming languages, such as TypeScript, JavaScript, Python, Java, Go, Rust, C++, C#, or Ruby.
- Ability to understand and modify unfamiliar codebases.
- A portfolio showing substantive authored code, not only documentation updates, dependency changes, formatting work, or minor corrections.
How it works
You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.
About AI training work
AI training work is the human work behind systems that learn from examples, including writing and reviewing code and rating model responses. Experienced software engineers are paid for this work because their judgment helps assess whether AI-generated code is correct, secure, complete, and maintainable.