Distributed Training Engineer, Sora

The Sora team at OpenAI is working on making video a key capability of OpenAI's foundation models. As a Distributed Training Engineer for Sora, you will work on improving the training throughput for our internal training framework and enable researchers to experiment with new ideas. This role requires strong engineering skills, the ability to write bug-free machine learning code, and deep knowledge of supercomputer performance.

Key responsibilities include:

• Collaborating with researchers to develop systems-efficient video models and architectures
• Applying the latest techniques to achieve impressive hardware efficiency for training runs
• Profiling and optimizing the training framework

The ideal candidate should have experience with multi-modal ML pipelines, strong software engineering skills (particularly in Python), experience with understanding and optimizing training kernels, and a passion for understanding stable training dynamics.

OpenAI offers a competitive compensation package, including a salary range of $295K – $440K, generous equity, and comprehensive benefits such as medical insurance, mental health support, 401(k) matching, unlimited time off, and paid parental leave.

This role is based in San Francisco, CA, with a hybrid work model of 3 days in the office per week. OpenAI is committed to diversity, equality, and creating an inclusive environment for all employees.