South Park Studios Company Logo

South Park Studios

Los Angeles, California - United States
View all jobs at this company »

Pipeline TD


Pipeline TD

At South Park Studios and DeepVoodoo, we’re looking for someone to help lay the tracks on a nascent pipeline that puts new and powerful computer vision and machine learning tools into the hands of iconoclastic creatives. As the Pipeline TD, you will be responsible for developing and maintaining software tools, providing support to artists, and general troubleshooting. The cement is still very wet here and the team is small, so we’re looking for someone to entrust with a lot of control over implementation details.

In this role you will

  • Develop pipeline and asset management tools to meet production demands

  • Interface directly with artists, engineers, and researchers and come up with elegant solutions to their problems

  • Have a high degree of autonomy and be expected to bring new ideas to the table

  • Provide front-line support for artists and production staff, troubleshoot assets and shots, and help with general problem-solving

  • Test and validate tools and techniques in production

  • Maintain documentation and provide hands-on training for other team members

We need you to have

  • 3-5 years relevant professional experience in a film, television, or commercial post-production environment

  • High degree of proficiency with Python, QT, JSON/XML/YAML, Git 

  • Familiarity with modern image processing and web development frameworks

  • Deep experience with production tracking database systems like Shotgrid or ftrack

  • Experience with render farms and job scheduling systems

  • Highly organized, motivated, self-starter mentality

We would be thrilled if you have

  • Passion for good user interface design

  • Understanding of artist workflows with industry standard DCC tools (e.g. from Foundry and Autodesk)

  • Familiarity with TensorFlow, PyTorch, and OpenCV

  • An interest in keeping abreast of the state of the art in computer vision and machine learning

  • Familiarity with systems administration

About Deep Voodoo

Founded by some of the most experienced and accomplished creatives in film and television, Deep Voodoo is a group of artists and engineers working to build new types of tools to make new types of art. We watched research in the fields of machine learning and computer vision deliver breakthroughs and finally grew tired of waiting for them to receive industry adoption. We formed 2 years ago with the goal of putting into practice exciting advances in image enhancement and matting, face tracking and swapping, and guided image synthesis. Now we’re ready for our next phase of growth.