AI Power User for Evaluating Advanced AI Systems

Join a pioneering AI research initiative focused on evaluating and improving how advanced AI systems perform complex, real-world personal workflows.

We are seeking experienced AI power users with extensive hands-on experience using Model Context Protocol (MCP) and AI plugins/connectors to complete multi-step personal tasks that require planning, judgment, context awareness, and tool integration.

This role is ideal for individuals who regularly rely on AI in their personal lives and understand how connected tools can enhance productivity across a wide range of everyday scenarios.

Requirements

• Design realistic prompts that reflect complex, real-world personal workflows.
• Execute multi-step tasks using AI while recording your screen as part of the evaluation process.
• Utilize personal AI plugins and connectors throughout workflow execution.
• Assess AI-generated outputs for practicality, personalization, reasoning quality, and overall usefulness.
• Identify where AI systems fail to understand context, misuse connected tools, overlook constraints, or produce unrealistic recommendations.
• Write detailed, structured explanations highlighting AI strengths, weaknesses, and improvement opportunities.
• Develop and apply comprehensive evaluation rubrics to measure AI performance consistently across different task types.

Areas of Focus

Your evaluations may include workflows involving:

• Personal health and wellness
• Travel planning and itinerary management
• Food, dining, and activity planning
• Home services and maintenance
• Career exploration and job search
• Productivity and personal organization
• Other complex day-to-day life management tasks

Required Qualifications

• Must be based in the United States.
• Strong hands-on experience with Model Context Protocol (MCP) and AI plugin/connector ecosystems.
• Frequent use of AI plugins/connectors such as Google Drive, Expedia, Notion, and similar productivity platforms.
• Heavy personal use of Large Language Models with an established usage history of at least 6 months.
• Experience using AI for multi-step planning, research, decision-making, and everyday personal workflows.
• Strong analytical thinking, written communication, and attention to detail.
• Ability to evaluate AI outputs for accuracy, usefulness, personalization, safety, and reasoning quality.
• Willingness to sign a data-sharing consent agreement through DocuSign.
• Experience designing, applying, or reviewing structured evaluation rubrics.

Preferred Qualifications

• Extensive rubric development and evaluation experience, including 100+ hours working on AI evaluation or quality assessment projects.
• Deep familiarity with AI-assisted productivity workflows involving multiple connected applications.
• Experience identifying AI limitations and recommending improvements based on real-world usage scenarios.

Why Join

• Help shape the future of intelligent AI assistants capable of supporting real-world personal workflows.
• Influence the development of AI systems that better understand context, preferences, trade-offs, and practical decision-making.
• Work on meaningful evaluations spanning health, travel, productivity, career planning, and everyday life management.
• Collaborate on cutting-edge AI initiatives with opportunities for continued engagement based on performance.

Engagement Details

• Expected commitment: 20+ hours per week.
• Initial onboarding and ramp-up period of approximately 1–2 days.
• Ability to complete assigned work within 24 hours of task assignment.
• Desktop or laptop computer required (Chromebooks are not supported).
• Selected contributors will begin with a trial period to evaluate quality, consistency, and overall project fit.
• As this project is in its early stages, there may be a short delay before initial task assignments become available.