Senior AI Networking Engineer - RoCEv2, NCCL, HPC

Xai; Memphis (TN); On-site; USD 180,000 - 240,000; Full time

Job summary

xAI is seeking an experienced engineer to design and operate large-scale ethernet AI/HPC networks. The role focuses on RoCEv2, NCCL usage, and building scalable GPU infrastructure for training and inference workloads. You will develop dashboards, optimize configurations, and own reliability across the fleet.

The position requires deep networking expertise, strong Python automation skills, and willingness to travel to Memphis for capacity expansion and on-call duties as the team grows.

Qualifications

• Designing and operating large scale networks in AI/HPC environments (10+ years).
• Expertise in Ethernet with Infiniband considered a bonus.
• Ability to train and troubleshoot AI training/inference workloads and NCCL usage.
• Develop a portfolio of metrics for performance and operations to optimize fleet.
• Automate tasks and analyze large datasets with Python.

Responsibilities

• Design and operate large scale network infrastructure for AI training/inference workloads.
• Build and tune metric dashboards to optimize fleet performance and availability.
• Design backend/frontend networks to enable scalable GPU infrastructure with minimal tooling.
• Travel to Memphis for capacity expansion and participate in on-call rotation; contribute to deployment/ops frameworks.

Skills

Large‑scale networks

Ethernet AI HPC

NCCL debugging

Performance metrics

Python automation

Network optimization

Job description

xAI is seeking an experienced engineer to design and operate large-scale ethernet AI/HPC networks. The role focuses on RoCEv2, NCCL usage, and building scalable GPU infrastructure for training and inference workloads. You will develop dashboards, optimize configurations, and own reliability across the fleet.

The position requires deep networking expertise, strong Python automation skills, and willingness to travel to Memphis for capacity expansion and on-call duties as the team grows.