Mercor
$150–190/hr
26 people hired
Advanced LLM power user with strong hands-on experience using MCP and plugins/connectors for real-world personal workflows. The role focuses on evaluating how effectively AI systems handle personalized, multi-step tasks requiring context, judgment, planning, and connected tools such as Google Drive, Expedia, Notion, and similar integrations. Experts will create realistic prompts, execute and screen-record workflows, use personal plugins/connectors, evaluate AI reasoning and tool usage, identify context or execution failures, and assess whether outputs are practical, personalized, safe, and realistic.
MCP, AI Plugins & Connectors, LLM Power User, AI Workflow Evaluation, Rubric Evaluation