Distributed Training Engineer, Sora

Details

The Sora team works on making video a core capability of OpenAI’s foundation models. This hybrid research-and-product team studies and expands video model capabilities while focusing on reliability and safety. The role focuses on improving training throughput for an internal training framework, enabling researchers to experiment with new ideas, and achieving high hardware efficiency for training runs. This position is based in San Francisco, CA and follows a hybrid work model (3 days in office per week). Relocation assistance is offered to new employees.

Responsibilities

• Collaborate with researchers to enable development of systems-efficient video models and architectures
• Apply the latest techniques to the internal training framework to achieve strong hardware efficiency for training runs
• Profile and optimize the training framework
• Design, implement, and optimize state-of-the-art AI models with robust engineering practices
• Write reliable, bug-free machine learning code and diagnose/fix issues impacting training performance

Requirements

• Strong software engineering skills and proficiency in Python
• Experience with distributed systems and an interest in understanding and improving their fundamentals
• Experience working with multi-modal ML pipelines (video/multi-modal models preferred)
• Experience understanding and optimizing training kernels and profiling/optimizing training performance
• Passion for stable training dynamics and improving hardware/training efficiency
• Ability to work closely with researchers and translate research needs into performant engineering solutions

Benefits

• Competitive base pay (see salary range)
• Equity and potential performance-related bonuses for eligible employees
• Medical, dental, and vision insurance with employer contributions to Health Savings Accounts
• Pre-tax accounts (Health FSA, Dependent Care FSA) and commuter benefits
• 401(k) retirement plan with employer match
• Paid parental leave, paid medical/caregiver leave, flexible PTO, and multiple paid holidays/company closures
• Mental health and wellness support; employer-paid basic life and disability coverage
• Annual learning and development stipend
• Daily meals in offices and meal delivery credits as eligible
• Relocation support for eligible employees

About OpenAI

OpenAI is an AI research and deployment company focused on ensuring general-purpose artificial intelligence benefits all of humanity. The company emphasizes safety, diverse perspectives, and inclusion. Background checks and reasonable accommodations for applicants with disabilities are addressed in accordance with applicable law and policy.