jobs

Software Engineer, Workload Enablement

About the TeamThe Scaling team is responsible for the architectural and engineering backbone of OpenAI’s infrastructure. We design and deliver advanced systems that support the deployment and operation of cutting-edge AI models. Our work sp

In inglese

Company

OpenAI

Location

San Francisco; Seattle

Status

Open

Posted

2026-03-28T00:39:41.241+00:00

OpenAI's Scaling team is hiring a software engineer to move production inference and training workloads onto new, sometimes early-access hardware. The work covers porting and validating those workloads, building end-to-end benchmarks and stress tests across CPU, GPU, memory, networking (NVLink, RDMA, WAN), storage and thermals, profiling distributed performance, and writing CI-runnable harnesses. It asks for a BS in CS or EE, 5+ years in ML systems, performance engineering, distributed systems or HPC, hands-on PyTorch, NCCL/RCCL and RDMA experience, and profiling skills; C++/CUDA/HIP is a plus. Vendor-facing debugging.