- United States +1 more
- Full-time
- 15+ years
Own the end-to-end network architecture for Runware's AI supercomputer, from scale‑up and scale‑out fabrics to storage, management, and inter‑site backbone, and lead the team building and operating the global network.
Locations (2)
- United States
- United Kingdom
Responsibilities
- Own the end‑to‑end network architecture for the AI Factory
- Design the optical layer and select transceivers, cables, and DWDM for pod‑to‑pod and metro links
- Define the reference network design for the Sonic Inference Pod
- Build and bring up the first fabrics, including rail‑optimized topologies, subnet management, routing, congestion control, QoS, and GPUDirect RDMA tuning
- Run capacity planning and vendor roadmap with NVIDIA, switch and optics suppliers, and carriers
Requirements
- Nice to have: NVIDIA Spectrum‑X
- Nice to have: Quantum‑X800
- Nice to have: Blackwell‑generation NVLink systems
- Nice to have: Co‑packaged optics