AI Workflow Evaluator & Plugin Expert (Remote)

United States Digital Space LLC; United States; Remote; USD 69,000 - 262,000; Full time

Job summary

Mercor seeks an MCP & Plug In Connectors Expert to evaluate AI systems on complex personal workflows and design realistic prompts using Google Drive, Expedia, Notion, and other plugins.

This remote, hourly contractor role requires strong MCP experience, heavy personal use of LLM products, an active LLM account with 6+ months, and willingness to sign a data-share consent form via DocuSign. US-based only; 20+ hours/week.

Qualifications

• US-based only.
• Strong MCP experience and plugin/connector usage.
• Experience using LLM plugins/connectors like Google Drive, Expedia, and Notion.
• Heavy personal usage of LLM products.
• Active LLM account with 6+ months of history.
• Willingness to sign a data-share consent form via DocuSign.
• Experience using AI for multi-step planning, research, and decision-making.
• Strong written judgment and attention to detail.
• Ability to evaluate AI outputs and write against rubrics.

Responsibilities

• Evaluate AI systems on complex personal workflows, including personal health, travel, and activity planning.
• Design realistic prompts for complex personal-life tasks using Google Drive, Expedia, Notion, and other plugins/connectors.
• Execute tasks while recording your screen and using personal plugins/connectors.
• Write clear explanations of AI successes and failures to improve AI model performance.
• Judge AI outputs for practicality, personalization, and reasoning.
• Identify where models miss context, overreach, or fail to use tools correctly.

Skills

MCP experience

Plugin/connector usage

LLM plugins (Drive/Expedia/Notion)

Personal LLM product usage

Active LLM account (>6 months)

Data-share consent experience

Multi-step planning

Written judgment

Rubric-based evaluation

Tools

Google Drive

Expedia

Notion

Job description

Mercor seeks an MCP & Plug In Connectors Expert to evaluate AI systems on complex personal workflows and design realistic prompts using Google Drive, Expedia, Notion, and other plugins.

This remote, hourly contractor role requires strong MCP experience, heavy personal use of LLM products, an active LLM account with 6+ months, and willingness to sign a data-share consent form via DocuSign. US-based only; 20+ hours/week.