AI Visual Brand Consistency Checker

Jorvex Studio

Remote · $160k – $200k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

A brand's visual identity is one of its most valuable assets. When an AI-generated marketing asset uses a font that's 2pt wrong, a blue that's 3 hex codes off from the brand standard, or a logo that's been placed too close to the edge of its safe zone, it erodes brand equity incrementally across millions of impressions. Manual brand compliance review doesn't scale to the content volumes that AI generation enables. Our platform generates marketing creative for 280 brands and produces 15,000 assets per week. Brand consistency checking is the quality gate between generation and publication. We have rule-based checks for simple constraints. We don't have a system that catches visual inconsistencies requiring perceptual understanding: layouts that feel off-brand even though every individual element is technically compliant, color harmonies that clash with established brand palette norms, or typography hierarchies that violate implicit but learnable brand conventions. You'll build the perceptual brand consistency checker: training vision models on brand-specific reference assets, implementing consistency scoring across visual dimensions, and building the feedback loop that teaches the checker from brand manager correction decisions. The checker needs to agree with brand managers at 90%+ rate to replace rather than augment manual review.

Fully remote.

Responsibilities

Design perceptual brand consistency models trained on brand-specific reference asset libraries
Implement multi-dimensional visual consistency scoring: color, typography, layout, and compositional feel
Build feedback learning system improving checker accuracy from brand manager correction decisions
Develop explanation outputs describing why a specific asset was flagged for inconsistency
Create consistency score calibration achieving 90%+ agreement with brand manager judgments
Build brand-specific model adaptation enabling rapid deployment for new brand clients

Requirements

6+ years in computer vision or ML engineering with visual quality, brand, or design system experience
Strong Python and PyTorch skills with perceptual quality assessment model development
Understanding of visual brand identity systems: color theory, typography hierarchy, and design principles
Experience training models on small-to-medium reference datasets with perceptual similarity objectives
Track record building vision systems that earn trust through demonstrated alignment with human judgment