This is a hybrid creative-engineering position for someone who thrives at the intersection of generative AI, software engineering, and animation workflows. You’ll prototype, train, deploy, and maintain new AI tools used in production by professional animators and creators. You’ll also work closely with the founding team to shape future directions in both research and product.
This is not a narrow ML role. You’ll fine-tune models, build production workflows, and handle aspects of MLOps. You’ll also do customer-facing work with studios and help draft specs for new capabilities. Expect to own the full lifecycle from research to MVP.
Requirements
- 2+ years in ML/AI engineering, with hands-on experience in diffusion models, transformers, generative image models, and building comfy workflows
- T-shaped: deep in ML, but comfortable shipping production systems - not just notebooks
- Experience fine-tuning and deploying models to production (LoRA, custom training pipelines, Docker, GPU infrastructure)
- Track record of building things from scratch - startups, side projects, open source, models you trained and shipped end-to-end
- Genuine interest in the creative/anime space - side projects, art, content creation, anything that shows intrinsic motivation beyond the paycheck
- English fluency
- Based in Japan or willing to relocate (visa sponsorship available)
Nice to haves
While not specifically required, tell us if you have any of the following.
- Japanese is a plus but not required
