Remote AI Image Policy Evaluator — Calibrate Models

Apply; Seattle, Northern (WA, KY); Hybrid; USD 62,000 - 76,000; Full time

Job summary

Handshake is seeking an AI Image Evaluator to assess prompts and generated images for quality, accuracy, and policy compliance. You will weigh visual details, compare outputs, and explain your reasoning to help train evaluation models.

Remote US role with flexible scheduling, part of Handshake AI's data-focused team. Prior experience with generative tools and content moderation is valuable, but a strong eye for detail can come from related fields.

Qualifications

• Ability to judge images and prompts with clarity and precision.
• Explain judgments using rubric language for auditability.
• Maintain consistency across many similar evaluations.
• Balance customer policy with project requirements.

Responsibilities

• Evaluate generated images against prompts for adherence and quality
• Compare images side by side and justify the selected result
• Classify imagery according to policy on sexual content, violence, IP, and minors
• Write concise rationales and cite rubric language
• Collaborate in calibration discussions and improve judgments over time

Skills

Visual analysis

Policy compliance

Rubric calibration

Clear writing

Critical thinking

Tools

Midjourney

Stable Diffusion

ComfyUI

Flux

DALL-E

Ideogram

Job description

Handshake is seeking an AI Image Evaluator to assess prompts and generated images for quality, accuracy, and policy compliance. You will weigh visual details, compare outputs, and explain your reasoning to help train evaluation models.

Remote US role with flexible scheduling, part of Handshake AI's data-focused team. Prior experience with generative tools and content moderation is valuable, but a strong eye for detail can come from related fields.