AI Model Evaluator — Visual Docs (Energy/Utilities) Remote

Obsidian; San Francisco (CA); Remote; USD 25,000 - 40,000; Part time

Job summary

Obsidian is seeking an independent contractor for a remote role focused on benchmark dataset projects that evaluate AI models on visual document understanding in the Energy/Utilities domain. The role requires authoring complex tasks and providing objective rubrics to assess performance.

The position is flexible, requiring approximately 15–20 hours per week, with competitive pay. Applicants from all backgrounds are welcome, with a focus on enhancing AI systems through human expertise.

Responsibilities

• Author complex, grounded tasks for AI evaluation.
• Provide objective rubrics for visual document understanding.
• Work in the Energy/Utilities domain.

Job description

Obsidian is seeking an independent contractor for a remote role focused on benchmark dataset projects that evaluate AI models on visual document understanding in the Energy/Utilities domain. The role requires authoring complex tasks and providing objective rubrics to assess performance.

The position is flexible, requiring approximately 15–20 hours per week, with competitive pay. Applicants from all backgrounds are welcome, with a focus on enhancing AI systems through human expertise.