Want to know if this job is worth applying to?
United States
Remote only
We're running a paid study on the quality and realism of coding environments designed to test AI agents. We are developing a comprehensive suite of programming tasks and evaluation harnesses to measure AI performance accurately. Your feedback directly shapes how these agents are benchmarked against real-world engineering standards.
During this remote session, you will review several programming tasks and their corresponding evaluation harnesses. You will assess the technical accuracy, complexity, and realism of each coding challenge. We will ask you to walk through the logic of the evaluation environments and identify any potential flaws. Finally, you will provide feedback on how these environments compare to standard industry practices.
We are looking for software engineers who have hands-on experience building, reviewing, or testing realistic programming tasks. We welcome full-stack developers, backend engineers, test automation engineers, and systems architects who are familiar with evaluation harnesses. Ideal candidates understand what makes a coding challenge robust, verifiable, and technically sound.
Review and assess the quality of realistic programming tasks
Evaluate coding environments and technical harnesses for AI testing
Walk us through potential flaws in the provided code structures
Provide feedback on the difficulty and realism of the challenges
Professional experience as a software engineer or developer
Familiarity with building or verifying programming tasks
Experience with code review and evaluation harnesses
Comfortable discussing technical architecture and testing methodologies
$75 per hour
Terac is building the world's largest pool of vetted human experts for AI. Researchers, AI labs, and product teams use Terac to recruit, screen, and pay study participants across industries, languages, and skill sets.
Learn more at terac.com or on YouTube at @jointerac.