About the Role
This role is central to building a serious, production-grade AI tooling capability as the generative AI space accelerates. You will own the deployment, operation, and development of our AI/ML tooling infrastructure, sitting at the intersection of cloud infrastructure, ML tooling, and VFX pipeline.
Responsibilities
- Provision and manage multi-GPU Ubuntu instances on AWS for AI/ML workloads
- Deploy and maintain services via systemd; manage cron jobs and scheduled tasks
- Bash scripting, environment management, dependency isolation
- Basic networking, reverse proxy configuration (nginx), port and access management
- Deploy and maintain ComfyUI as a shared studio service; develop and manage custom nodes
- Multi-GPU load balancing across concurrent artist users
- Build and deploy Gradio applications as production-grade internal tools
- Manage diffusion model workflows across image, video and 3D domains
- HuggingFace model management including gated model access
- LoRA training and model fine-tuning for studio-specific use cases
- Python scripting for pipeline automation and tooling integration
- REST API development and integration (Flask / FastAPI / Gradio)
- Git-based version control and workflow management
- Understand VFX production context: Nuke, EXR, colour pipelines
- Develop and maintain AI tooling for VFX-specific use cases: roto, upscaling, comp integration
- Monitor and maintain GPU utilisation across concurrent production users
- Document services and own the onboarding experience for non-technical studio users
- Continuously evaluate and integrate new open-source models and tools relevant to VFX production
Requirements
- Significant experience working in a core Technology team
- In-depth knowledge of editorial, post production and VFX workflows
- Knowledge of Render Farms and Render Management Software (Deadline in particular)
- Worked with shared storage solutions and are confident handling critical and sensitive company data
- Knowledge of a range of operating systems including Linux, OSX and Windows
- Proficient in Python and Bash scripting
- Comfortable spinning up multi-GPU Ubuntu instances on AWS, building ComfyUI workflows with custom nodes, and handing off polished tools to artists who don’t know what a conda environment is.
- Understand enough about the production context to make decisions about what gets built, how it gets deployed, and what the artist experience should look like.
- Highly motivated problem solving individual
- Thrive in leading a close knit team to find solutions to complex Technology challenges
- Have great ideas and the skills to bring the best ideas from the wider team
- Experience ensures that their implementation is done in the most comprehensive way.
Skills
- AWS
- Linux
- Bash scripting
- ComfyUI
- Gradio
- HuggingFace
- Python
- Git
- Nuke
- EXR
- Flask
- FastAPI
- nginx
- systemd
- cron jobs
- REST API
- Cloud technology
- Deployment technologies
- Orchestration technologies
Location
- Remote
Work Type
- Full-time
Experience Level
- Mid-level
- Senior
Benefits
- Ongoing development support
- Personal progression plan
About the Company
- Untold Studios is a creative community and studio focused on creating outstanding content that captivates audiences. We operate entirely in the cloud across Commercials, Feature Films, and Episodic work for major streamers.
- Our people and culture are at the core of everything we do.
- Our purpose is to do brilliant creative work, to make content that audiences want to watch, and to have fun while doing it.
