Principal AI Infrastructure Network Engineer (Equity)

Nscale; Seattle, New York, San Francisco, Houston (WA, NY, CA, TX); Sur place; USD 180 000 - 240 000; Plein temps

Avantages offerts par ce poste

Base salary + equity

Equity incentives

Dynamic progression plan

Résumé du poste

Nscale in Seattle is seeking a Principal Network Engineer – AI Infrastructure to provide technical leadership across high-speed networking, focusing on Infiniband and RDMA fabrics. You will own reliability, scalability, and long-term evolution, guiding architecture and standards across teams.

You’ll mentor engineers, drive cross-team initiatives to improve uptime and latency, define hardware configurations and routing practices, and influence end-to-end system design with SRE and compute

Qualifications

• 10+ years in network engineering with focus on HPC/AI or hyperscale data centers.
• Expertise in Infiniband and large-scale RoCE fabrics.
• Understanding of RDMA internals and congestion management.
• Proficiency in data center routing/control planes (BGP/OSPF/ECMP).
• Ability to lead complex initiatives across teams without direct authority.

Responsabilités

• Own technical direction and operational strategy for AI interconnect networks.
• Design and evolve large-scale Infiniband/RoCE fabrics to support growth.
• Lead complex investigations and provide systemic fixes for incidents.
• Drive cross-team efforts to improve fabric reliability and performance.
• Define standards for hardware config, congestion control, and routing.

Connaissances

HPC networking

RoCE fabric

RDMA internals

BGP/OSPF/ECMP

Cross-team leadership

Outils

Mellanox NICs

Description du poste

Nscale in Seattle is seeking a Principal Network Engineer – AI Infrastructure to provide technical leadership across high-speed networking, focusing on Infiniband and RDMA fabrics. You will own reliability, scalability, and long-term evolution, guiding architecture and standards across teams.

You’ll mentor engineers, drive cross-team initiatives to improve uptime and latency, define hardware configurations and routing practices, and influence end-to-end system design with SRE and compute