SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Highrise AI, a GPU-native infrastructure platform subsidiary of Hut 8 Corp., is seeking a Principal AI Network Engineer to lead the design, deployment, and optimization of large-scale GPU cluster networks. The role bridges physical installation and cabling through full production operation, working alongside senior compute, storage, and infrastructure engineers.
You will own network fabric bring-up for AI clusters, including switch initialization, configuration validation, and host connectivity readiness. Key responsibilities include configuring high-speed Ethernet and InfiniBand compute fabrics with RDMA/RoCE support, validating topology, routing, redundancy, addressing, port mappings, optics, and cabling across switches and NICs. You will implement network automation, configuration management, and repeatable fabric deployment procedures, partnering with compute and storage teams to resolve integration issues.
Performance optimization is central to the role. You will execute fabric qualification and acceptance testing for link health, error rates, bandwidth, latency, congestion, and resilience. You will troubleshoot and tune RDMA transport, congestion control, QoS, PFC/ECN, routing, and switch/NIC settings. Working with compute engineers, you will analyze NCCL and distributed workload results to isolate communication bottlenecks and improve multi-node performance.
Following deployment, you will serve as a Tier 3 escalation point for complex production networking issues, driving root cause analysis, reliability improvements, and ongoing optimization of high-performance AI clusters. The role is hands-on and technical, requiring deep expertise in cluster networking at scale.
Highrise combines bare-metal performance with full-stack orchestration, API-driven access, and hardware-enforced isolation to support training, fine-tuning, and production-scale AI model deployment. The company operates both Highrise-owned GPU fleets and third-party hardware on behalf of customers and partners, applying consistent operational standards across all infrastructure.
REQUIREMENTS:
The posting excerpt provided does not include an explicit requirements section. Candidates should have demonstrated expertise in large-scale network engineering, GPU cluster infrastructure, high-speed Ethernet/InfiniBand configuration, RDMA/RoCE protocols, network automation and orchestration, and production troubleshooting of complex distributed systems. Principal-level candidates are expected to have 10+ years of relevant infrastructure or network engineering experience, with proven success leading technical initiatives in high-performance computing or cloud infrastructure environments.
