Training Performance Engineer

OpenAI is seeking a Training Performance Engineer to improve efficiency and optimization across their distributed machine learning training run stack. This role involves analyzing training data, optimizing GPU utilization, and implementing improvements to improve the performance of training jobs. The Engineer will work in a hybrid model of three days in the office per week.

Requirements

• Love optimizing performance and digging into systems to understand how every layer interacts.
• Strong programming skills in Python and C++ (Rust or CUDA a plus).
• Experience running distributed training jobs on multi-GPU systems or HPC clusters.
• Enjoy debugging complex distributed systems and measuring efficiency rigorously.
• Comfortable collaborating across teams and translating raw profiling data into practical engineering improvements.