AI Consistent Character Style Transfer Developer

Elvio Works

Remote · $130k – $160k / 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

Generating a single stylized image of a character is a solved problem. Generating 200 images of the same character — same face, same proportions, same distinctive features — across different poses, expressions, lighting conditions, and environments is not. Identity consistency at scale is the frontier that separates generative AI demos from production creative tools. Our platform serves 2,400 game studios and animation production companies who need consistent character assets across hundreds or thousands of frames. The consistency problem has several dimensions: facial identity preservation, style consistency across different scene lighting, and maintaining distinctive non-facial features like scars, accessories, and clothing details. You'll own the consistency layer of our generation pipeline — reference conditioning mechanisms, automated consistency scoring that detects drift before it ships, and the fine-tuning pipelines that let clients train the system on their existing character designs.

Fully remote.

Responsibilities

Design reference conditioning mechanisms for identity-consistent character generation
Build automated consistency scoring to detect character drift across a generation batch
Develop client-facing fine-tuning pipelines for custom character training
Implement consistency metrics aligned with client quality standards
Collaborate with generative model researchers on architecture improvements
Optimize generation throughput for large-batch character asset production

Requirements

5+ years in computer vision or generative model engineering
Experience with diffusion models, ControlNet, or reference conditioning techniques
Strong PyTorch skills and GPU programming knowledge
Understanding of identity preservation challenges in generative models
Portfolio demonstrating generative image quality and style consistency work