Netflix lists ComfyUI in job postings for INKubator animation studio

Netflix is building INKubator, an experimental animation studio that will produce AI-assisted short films before scaling to features. Job postings for CG Experimental Artists list ComfyUI as a required tool alongside Photoshop, Maya, Houdini, Harmony, and Blender — specifically for generating concept frames, style frames, character explorations, and environment designs from scripts or verbal direction. One listing also asks candidates to "build and fine-tune generative image and video models."

The ComfyUI requirement marks the first time a major studio has named the open-source node-based diffusion tool in a job posting at this level. ComfyUI has been the de facto standard for local image generation workflows among practitioners since late 2023, but its appearance in a Netflix hiring document signals that generative pipelines are moving from hobbyist experiments to production infrastructure. The studio is treating ComfyUI as peer software to Maya and Houdini — commercial packages that cost thousands of dollars per seat and have defined VFX and animation pipelines for decades.

INKubator is linked to Netflix's $600 million acquisition of InterPositive, Ben Affleck's AI filmmaking startup, announced earlier this year. Serrena Iyer, formerly of DreamWorks Animation, A24, and MRC, is among the LinkedIn profiles associated with the studio. The job postings frame generative tools as one component in a hybrid pipeline: artists use ComfyUI for rapid concept iteration, then hand off to traditional animation software for refinement and final production. The model-fine-tuning requirement suggests Netflix intends to train custom checkpoints on its own IP rather than rely on public Stable Diffusion or FLUX weights — an approach other studios have quietly adopted over the past year, using open-source tooling as scaffolding while locking down weights and workflows internally.