The work
You will help evaluate software engineering work used to train large language models. The role combines code review, solution development, error analysis, and clear written explanations of your decisions.
You will work across backend and frontend codebases and help create reliable ways to check whether software solutions meet performance and quality standards.
- Curate code examples and develop precise software solutions.
- Correct implementations and review AI-generated code.
- Evaluate efficiency, scalability, reliability, and code quality.
- Work with Python, JavaScript, React, Node.js, C/C++, Java, Rust, and Go.
- Identify error patterns and help build mechanisms that verify software solutions.
- Write clear evaluation rationales against relevant performance benchmarks.
What it pays and takes
The source does not provide a pay rate for this role. It is a remote contractor engagement for candidates based in the United States, with a flexible commitment and an initial one-month duration.
Strong practical software engineering ability matters most. Comparable experience and strong computer science foundations are relevant, while exceptional skills take precedence over educational background.
- Location: United States.
- Commitment: 10 to 40 hours per week.
- Initial duration: One month, with potential extensions based on performance and fit.
- Experience: At least three years of software engineering experience.
- Skills: Full-stack development with Python and JavaScript, including React and Node.js.
- Production experience: Building and deploying scalable, production-grade software.
- Knowledge: Software architecture, design, development, debugging, and code review across backend and frontend codebases.
- Communication: Fluent in English with strong oral and written communication skills.