Algorithms and Theory Researcher (Research Coding Tasks)

You will create self-contained technical tasks in algorithms and theory to help evaluate AI models. You will also develop or use reference implementations, which show a correct and usable solution, and review model outputs for correctness, rigor, and research quality.

The work covers areas such as graph, approximation, streaming, and online algorithms. You will work asynchronously with a project lead to clarify task scope, maintain a high technical standard, and provide precise feedback that connects theory with working code.

• Build research coding tasks based on your algorithms research area.
• Develop or work with reference implementations for algorithms problems.
• Review AI model outputs for algorithmic reasoning, correctness, and implementation quality.
• Complete self-contained tasks through an asynchronous workflow that can fit around ongoing research.

What it pays and takes

This role is suited to researchers and research engineers who publish their work and release usable research code.

• Pay: $200 per approved task.
• Task length: Each self-contained task takes about four hours.
• Work arrangement: Fully remote, asynchronous contract work.
• Location: Open to candidates in the countries specified for this role.
• Language: Professional fluency in English.
• Required: A completed or in-progress PhD in computer science or a related field.
• Required: Published research in algorithms at venues such as SODA, STOC, FOCS, ESA, ICALP, or comparable venues.
• Required: Research code released through GitHub or another accessible research artifact.
• Required: Ability to assess algorithmic reasoning, correctness, and implementation quality.
• Helpful: First-author publications and depth in graph, approximation, streaming, or online algorithms.
• Helpful: Experience with Docker or reproducible research environments.

How it works

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

About AI training work

OpenTrain is a platform where people find AI training and data-labeling projects, build a profile, and apply. AI training work uses human-written examples, reviews, and evaluations to help improve artificial intelligence systems.