The work
You will create realistic backend coding problems used to train and evaluate AI systems. The work covers software bugs, new features, refactoring, and performance improvements.
You will also build reproducible environments, deterministic checks, and reference solutions so valid code is accepted and incorrect code is rejected. You will review technical submissions and explain your technical reasoning clearly.
- Create coding tasks for feature implementation, bug fixing, refactoring, and performance optimization.
- Build reproducible environments and deterministic verifiers for coding problems.
- Produce golden reference solutions for each task.
- Analyze complex defects and performance bottlenecks in large or legacy codebases.
- Implement scalable, maintainable features and document solution approaches for AI training data.
- Review technical submissions for accuracy and clarity.
What it pays and takes
This is an entry-level, part-time contractor role with less than 20 hours per week. The work is open to candidates in the countries listed below and requires professional fluency in English.
- Pay: $50 to $100 per hour equivalent, paid per task that meets the project specifications.
- Hours: Less than 20 hours per week.
- Location: Open to candidates in the United Arab Emirates, Argentina, Austria, Bangladesh, Belgium, Bahrain, Brazil, Canada, Switzerland, Chile, Colombia, Germany, Denmark, Algeria, Egypt, Spain, France, the United Kingdom, Ghana, Greece, Indonesia, India, Italy, Jordan, Japan, Kenya, Kyrgyzstan, Kuwait, Kazakhstan, Lebanon, Sri Lanka, Morocco, Mexico, Malaysia, Nigeria, the Netherlands, Oman, Peru, the Philippines, Pakistan, Palestine, Portugal, Qatar, Saudi Arabia, Singapore, Thailand, Tunisia, Taiwan, the United States, Uzbekistan, Vietnam, and South Africa.
- Language: Professional fluency in English.
- Programming: Hands-on expertise in Python, Java, Rust, Go, C++, TypeScript, Node.js, or a similar backend language.
- Engineering: Experience building scalable services and APIs, debugging software, writing automated tests, and applying algorithms and data structures.
- Code quality: Experience resolving performance bottlenecks, refactoring large or legacy codebases, reviewing code, documenting technical decisions, and communicating clearly.
- Portfolio: Clear open-source contributions or an equivalent portfolio that demonstrates coding ability.
- Helpful background: Experience delivering backend features across cross-functional projects, strong public GitHub work, and an interest in AI systems. Previous AI experience is not required.
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 is the human work behind systems that learn from examples. In this role, software engineers create and review coding tasks, solutions, and checks that help AI models learn to write and improve code; experienced contributors are valued for their technical judgment and domain knowledge.