Network Engineer, Supercomputing

Thinking Machines Lab is seeking a Network Engineer to oversee the foundational network infrastructure supporting large-scale training and inference operations. This role focuses on ensuring the reliability and performance of GPU interconnects, including RDMA/RoCE fabrics between nodes and NVLink/NVSwitch domains within GPUs. The engineer will be responsible for diagnosing and resolving network issues that can impact training processes, ensuring seamless connectivity across extensive GPU fabrics.