← Back to vacancies

AI Image Evaluation Analyst

Japan, France
By agreement
Japan, France (Remote)RemoteContractiMerit TechnologyNEW

Description (EN)

The work

iMerit, an EXL company, is looking for detail oriented analysts to evaluate and rank AI generated responses to image based prompts. You will review AI generated responses and decide which ones actually hold up: accurate, relevant, concise, safe, well localized, and true to what the user asked for. Every ranking comes with a short written rationale, and those comments feed directly into how the models improve.

Careful reading and good judgment matter far more than speed here.

What you will do

  • Interpret conversational context and identify what the user really wanted

  • Rate and rank responses against defined quality criteria

  • Compare multiple answers and explain in writing why one wins

  • Verify factual claims using approved research sources

  • Flag tasks that cannot be reliably assessed rather than guessing

What you bring

  • Strong critical thinking and sound judgment in ambiguous cases

  • Solid research skills and attention to detail

  • Excellent reading comprehension

  • Professional written English that is clear and concise

  • Self direction and the discipline to hit deadlines without supervision

Prior experience in AI evaluation, annotation, writing, editing, research, or translation is a plus, though not required.

Good to know

  • Independent contractor engagement for the length of the project. 

  • Fully remote and flexible. You choose your hours as long as volume and deadlines are met.

  • Task volume varies with project demand.

  • A paid qualification assessment is required before live work begins.

 

Published 1 day ago

Similar vacancies

By agreement
Paris, FranceFull-timeLeaddatabricks

FEQ427R248 As a Field Engineering Manager, you will lead a team of technical pre-sales Solutions Architects and Solutions Engineers across the UAE mar...

By agreement
Paris, FranceFull-timeSeniordatadog

AI Platform builds the foundations of Datadog's AI efforts. The org is 70+ people organised in three pillars: training and serving (GPU clusters, dist...

By agreement
Paris, FranceFull-timeLeadAnthropic

About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users...

By agreement
Paris, FranceFull-timeSeniordatabricks

CSQ427R154 As a Forward Deployed Engineer (FDE) you will work with customers to build and productionize solutions to their data & AI challenges us...