Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.
MCP & Plug In Connectors Expert
United States
$50 – $190/hr
July 16, 2026
Job description
Role Overview
A leading AI research organization is seeking advanced LLM power users with deep experience using Model Context Protocol (MCP) and AI plugins/connectors to evaluate how well AI systems perform real-world, multi-step personal workflows.
You will assess AI systems on practical tasks involving connected tools such as Google Drive, Expedia, Notion, and similar integrations. The role focuses on evaluating personalized AI assistance across planning, decision-making, and life organization rather than simply producing AI-generated outputs.
This is a flexible remote contract opportunity for experienced AI power users who regularly use MCP and connected tools in their daily lives.
Key Responsibilities
Evaluate Real-World AI Workflows
- Test AI systems across workflows involving:
- Personal health
- Travel planning
- Food and dining
- Home services
- Career search
- Personal organization
- Other real-world life management tasks
Create & Execute Evaluation Tasks
- Create realistic prompts for complex personal-life scenarios.
- Execute multi-step workflows while recording your screen.
- Use personal MCP servers and AI plugins/connectors throughout task execution.
Assess AI Performance
Evaluate AI responses for:
- Practicality
- Personalization
- Context awareness
- Reasoning quality
- Tool usage
- Safety
- Real-world usefulness
Identify:
- Missing context
- Poor judgment
- Incorrect connector usage
- Unrealistic recommendations
- Hallucinations
- Workflow failures
Develop Evaluation Rubrics
- Design detailed scoring rubrics for AI evaluations.
- Apply evaluation criteria consistently across multiple tasks.
- Write structured explanations describing AI successes, failures, and improvement opportunities.
Required Qualifications
Must be based in the United States.
Strong experience using Model Context Protocol (MCP).
Extensive experience using AI plugins/connectors such as:
- Google Drive
- Expedia
- Notion
- Similar connected applications
Heavy personal use of LLM products.
Active AI account with approximately 6+ months of regular usage history.
Willingness to sign a data-sharing consent form via DocuSign.
Experience using AI for:
- Planning
- Research
- Decision-making
- Personal productivity
- Multi-step workflows
Strong written communication and analytical skills.
Ability to evaluate AI outputs for quality, completeness, realism, and safety.
Experience creating and applying evaluation rubrics.
Preferred Qualifications
100+ hours of prior rubric development, evaluation, or quality assessment experience.
Extensive experience evaluating AI systems or RLHF projects.
Compensation
- $50–$190 USD per hour
Work Arrangement
- Remote
- Independent contractor
- Approximately 20+ hours per week
- Flexible schedule
- Trial period before ongoing task assignment
- Desktop or laptop required (Chromebooks not supported)
- Weekly payments via Stripe or Wise
Apply now
You will be redirected to the company's website to complete your application.
Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.
Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.
Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.