Benchmark Author

Pathwize is looking for Machine Learning professionals to support frontier AI research through flexible, hourly contract work. This is not a traditional job. You'll draw on your hands-on experience in machine learning to evaluate AI-generated work and supply the expert judgment that helps models reason about machine learning the way a credentialed specialist would.

Every contribution is captured with a verifiable, EU-native provenance trail, so the data you produce is auditable by design. This is an ongoing, project-based opportunity that can be done alongside research, teaching, postdoctoral work, or industry employment.

• You'll complete real, expert-level tasks drawn from your professional practice, structured as discrete projects with clear instructions and a transparent record of what the AI drafted versus what you decided.
• Fully remote, with a flexible and asynchronous schedule and no minimum-hour requirement; most contributors work roughly 5 to 20 hours per week when participating in an active project.
• Placement depends on current project needs, with opportunities to be considered for future projects as they become available.

• At least 2 years of professional or research experience in machine learning or a closely related field.
• Able to independently produce rigorous, well-documented work and to reason clearly about edge cases.
• Education: one of Bachelor's, Master's, Ph.Ds.
• Strong written communication, attention to detail, and comfort working independently and asynchronously.
• Right to work in the EU or EEA (Pathwize is EU-native; all data stays in the EU).