Networking Software Engineer (C++/CUDA, RDMA)

OpenAI; California (MO); Hybrid; USD 180,000 - 250,000; Full time

Benefits offered by this job

Relocation assistance

Job summary

OpenAI is seeking a Software Engineer, Networking in San Francisco to design and implement custom networking collectives tightly integrated into our training stack. You will work on low-level, performance-critical code in C++ and CUDA, collaborating with ML researchers to ensure efficient communication across our largest training jobs.

The role offers a hybrid work model (3 days in the office per week) and relocation assistance for new employees.

Qualifications

• Experience with low-level performance-critical software.
• Background in distributed algorithms is a plus.
• Experience with collective communication is a bonus.

Responsibilities

• Collaborate with ML researchers to design and implement efficient collective operations in C++ and CUDA.
• Ensure largest training jobs take full advantage of network transports used in our supercomputers.
• Develop simulations to inform future supercomputer network designs.

Skills

C++

CUDA

Distributed algorithms

RDMA

Low-level performance

Job description

OpenAI is seeking a Software Engineer, Networking in San Francisco to design and implement custom networking collectives tightly integrated into our training stack. You will work on low-level, performance-critical code in C++ and CUDA, collaborating with ML researchers to ensure efficient communication across our largest training jobs.

The role offers a hybrid work model (3 days in the office per week) and relocation assistance for new employees.