Design and maintain complex workflows in ComfyUI, integrating cutting-edge research to solve production challenges.
Generate final, screen-ready character elements for a feature film in production, ensuring deliverables meet strict artistic standards within the standard review and approval cycle.
Lead the exploration and implementation of open-source video generation models focusing on maintaining temporal consistency, high-fidelity motion, and integrating image-to-video and video-to-video workflows.
Utilize recent advancements in instruction-based editing and accelerated inference models (turbo/distilled architectures) to allow for rapid iteration and precise, prompt-driven manipulation of existing assets.
Leverage conditioning tools (e.g., ControlNet, IP-Adapter) to shape and define the output of AI-generated media, ensuring precise control over composition, pose, and style.
Manage workflows for training and implementing adapters (LoRA) and embeddings to generate consistent results that fit the specific artistic vision of the project.
Prepare and refine AI-generated assets in Photoshop, utilizing traditional digital art techniques to correct artifacts and composite layers for the highest quality finish.
Work closely with artists and production teams to integrate generative AI techniques into the overall pipeline, ensuring smooth workflows for photorealistic content.

Advanced proficiency with ComfyUI and Python scripting for custom node creation is highly preferred. Ability to troubleshoot and optimize complex node graphs.
At least 5 years of professional experience in a technical creative role (e.g., Technical Artist, Creative Technologist, or Digital Designer) demonstrating a mastery of high-fidelity, feature-film quality digital art workflows and technical problem-solving.
Deep understanding of current neural network architectures, including Diffusion Transformers (DiT) and Flow Matching. Experience working with open-source video models and large-scale image editing models is essential.
Advanced skills and knowledge using Photoshop to refine photorealistic imagery and prepare input assets.
Experience managing compatibility between various architecture types, ensuring the correct use of VAEs, text encoders, and floating-point precision standards when transitioning between different generations of open-source models.
A strong portfolio that demonstrates your ability to combine technical expertise and artistic vision using generative AI techniques for both static and moving images.

Technical Artist – Generative AI R&D

Key skills