Image Annotation QA
Pay needs confirmation
Remote · Country eligibility: check the official listing · Work type not specified
Loading opportunities…
Per task · confirm rate per accepted task
We are seeking detail-oriented QA Reviewers to check and approve image edits used to train next-generation AI image editing models.
Listed location: Remote · Country eligibility: check the official listing
Find roles for my country and experience →Provider description (Turing)
About Turing
Turing is an AI-powered tech services company with a mission to accelerate AGI advancement and deployment by bridging the gap between global talent from 100+ countries and the world's best foundational LLM companies, by helping them improve performance through model evaluation, fine-tuning, feedback, factuality and handling diverse data.
Role Overview
We are seeking detail-oriented QA Reviewers to check and approve image edits used to train next-generation AI image editing models. Each task contains an original image, a written editing instruction (prompt), an edit category (L2) and the edited image produced by an annotator. You verify that the edit follows the instruction exactly, looks photorealistic and leaves everything else untouched, then approve it, fix it or send it back with clear feedback.
This role suits senior photo retouchers, art directors, creative leads and image QA specialists who combine a sharp eye for visual defects with strong written English. It is a remote, project-based role, with training and guidelines provided.
What does day-to-day look like:
Review edited images against the original image, the prompt and its L2 category, and confirm the edit does exactly what was asked, nothing more and nothing less
Check realism and technical quality: lighting, shadows, reflections, perspective, edges, color consistency, and no halos, seams, warping or artifacts
Confirm that areas the instruction does not mention are unchanged, including identity, pose, background and resolution
Score each task against the project quality rubric and mark it as approved, minor fix or rework
Write constructive, specific feedback and detailed annotations so annotators know exactly what to correct
Flag prompts that are unclear, impossible or inconsistent with their L2 category, and escalate recurring issues to the project lead
Spot error patterns across annotators and contribute to guideline updates and calibration sessions
Meet daily review throughput and accuracy targets while keeping scoring consistent with other reviewers
Requirements:
English Proficiency:
Ability to read and write in English with a high degree of comprehension
Analytical Skills:
Good research and analytical skills to judge edits objectively against written guidelines
Feedback Skills:
Ability to provide constructive feedback and detailed annotations
Creative Thinking:
Creative and lateral thinking to judge ambiguous prompts and suggest workable fixes
Communication:
Excellent communication and collaboration skills for working with annotators and project leads
Independence:
Self-motivated and able to work independently in a remote setting
Technical Setup:
Desktop or laptop with a color-accurate display and a good internet connection
Experience: 1
+ years in photo retouching, compositing, art direction, creative QA or a similar role
Visual Expertise:
Strong eye for photorealism and the ability to spot subtle editing defects at full resolution
Tools:
Working knowledge of Adobe Photoshop, enough to verify edits and make small corrections
Good to have:
Prior experience reviewing or managing a team of retouchers or designers
Agency or in-house studio experience in fashion, beauty, FMCG, e-commerce or retail
Prior data annotation, content moderation or QA work against a scoring rubric
Familiarity with AI image tools (Photoshop Generative Fill, Firefly, Midjourney, Stable Diffusion)
Benefits
Opportunity to work on cutting-edge AI projects.
Competitive compensation.
Flexible working hours and remote work environment.
About the project
Its a pay per task
Duration is 12 weeks
Source: Turing official referral page, checked 2026-11-12. An application entry was present; hiring capacity and final terms remain subject to Turing.
The listing states Per task · confirm rate per accepted task. Confirm the final rate, workload, and payment terms during the official application process.
Use your CV to find relevant roles on HumanitApp. This is separate from a partner application.
Match my CV →Preview the latest role picks before choosing weekly email updates.
Read this week’s picks → · Get weekly role alerts →Before you apply
The listing states Per task · confirm rate per accepted task. Confirm the final rate, workload, and payment terms during the official application process.
Yes. The button opens the exact verified Turinglisting using the referral URL published for this role.
No. HumanitApp independently curates the opportunity. The partner platform manages applications and hiring decisions.
No. Choose the primary Apply action to continue directly without giving HumanitApp your name or email address.