Practicing software engineers review coding tasks and evaluation harnesses used to benchmark AI agents, checking that they reflect real engineering problems. Done in a remote, guided session with screen sharing; open worldwide, $65 per hour.
Experienced software engineers based in South Asia walk through proposed coding tasks and environments in an AI-moderated session, giving feedback on realism, difficulty and test suites used to evaluate AI coding agents. Remote, $40 per hour.
Software engineers based in Brazil or Argentina review proposed coding tasks and evaluation harnesses used to test AI agents, sharing their screen and giving feedback on flaws. Terac lists USD 40 per hour.
Experienced software engineers design realistic full-stack coding tasks and the tests that grade them, used to train and evaluate AI coding agents. Remote contract, $45–75/hour paid weekly, 10–20+ hours a week, with a paid work trial.
New roles land most days. Get them on Telegram or Discord as they are added, or read how the platforms pay before you apply.