AI Interaction Evaluator (Codex / Claude Code, up to $200/hr)
G2i
Miami
Workplace: RemoteContractUSD 50 - 200 hourlyFunction: People, HR & TalentSkills: ["Subjective judgment","Rigorous evaluation","Direct feedback"]Evaluate the end-to-end quality of AI coding agent interactions in real-world scenarios. You’ll judge whether responses make sense, whether preambles and reasoning are useful, and whether outputs reflect strong engineering judgment and “taste.” This is not production coding work—it's assessing explanation quality, trust-building, and how well the agent guides users, including using tools like Cursor, Codex, and Claude Code.

