R

Runware

Generative AI platform with fast inference

AI Factory Network Architect

Network Engineering & AdministrationUpdated on Nov 11
Full-timeRemote in USA

About the Job

Requirements

  • At least 15 years of experience designing, building, and operating high-performance networks, with significant experience in HPC or AI supercomputers at the scale of thousands of GPUs or more.
  • Deep hands-on InfiniBand experience in fabric design, subnet management, routing engines, adaptive routing, congestion control, and performance debugging at the port and queue-pair level.
  • Working knowledge of NVLink and NVSwitch domains and how scale-up and scale-out fabrics interact for multi-GPU and multi-node inference and training.
  • Optical networking expertise, including 400G and 800G transceivers and cabling, link budgets, DWDM and coherent optics for inter-site links, and operational issues such as failure rates, cleaning, and inventory.
  • Hands-on RDMA experience across InfiniBand and RoCE, including GPUDirect, NCCL and collective behavior, PFC and ECN tuning, and the trade-offs between InfiniBand and RoCE.
  • A track record of leading network teams through greenfield builds and production incidents, and serving as the escalation point for complex fabric issues.
  • Ability to automate using Python or Go, infrastructure as code, and telemetry pipelines.
  • Clear writing and the ability to create documentation that site teams in other locations can execute.
  • Willingness to travel to pod sites in the US and Europe for bring-up and major changes.

Responsibilities

  • Own end-to-end network architecture for the AI Factory, including scale-up NVLink and NVSwitch domains, scale-out InfiniBand and RoCE fabrics, storage and management networks, and the pod-to-pod and inter-site backbone.
  • Design the optical layer by selecting transceivers and cables, evaluating co-packaged and linear-drive optics, designing DWDM for pod-to-pod and metro links, and setting link budgets and reliability targets.
  • Define a reference network design for the Sonic Inference Pod so each new pod ships with a validated, repeatable fabric and joins the global network in days.
  • Build and bring up the initial fabrics, including rail-optimized topologies, subnet manager and routing configuration, adaptive routing, congestion control, quality of service, and GPUDirect RDMA tuning; measure performance against inference throughput and tail latency.
  • Work with the inference engine team so traffic patterns—including tensor- and pipeline-parallel collectives, KV-cache transfer, and multi-node serving—inform topology and buffer decisions.
  • Run capacity planning and the vendor roadmap with NVIDIA, switch and optics suppliers, and carriers; negotiate technical specifications and lead evaluations of new fabric generations.
  • Set operating standards for telemetry, link-health monitoring, failure-domain design, automated fault isolation, and runbooks for operating tens of thousands of ports.
  • Hire and lead a team of network engineers as the footprint grows, and serve as the final technical escalation for fabric incidents across all sites.

Desired Qualifications

  • Experience with NVIDIA Spectrum-X, Quantum-X800, Blackwell-generation NVLink systems, or equivalent hyperscaler in-house stacks.
  • Hands-on experience with co-packaged optics, linear-drive pluggables, or photonic switching evaluations.
  • Experience designing networks for modular or containerized data centers, edge sites, or deployments with constrained power and space.
  • Experience with inference-specific traffic patterns, including disaggregated prefill and decode, KV-cache transfer, and multi-tenant serving across a shared fabric.
  • Contributions to SONiC, OpenFabrics, UEC, or similar open networking efforts.
  • Experience with sovereign or data-residency network designs, including inter-site encryption and segmentation.
  • Prior work at a GPU cloud, hyperscaler, national lab, or AI lab building its own supercomputer.

About the Company

Runware

Generative AI platform with fast inference

Runware.ai provides a fast, cost-effective platform to deploy AI-generated content through an API that hides infrastructure complexity. It runs AI workloads on custom hardware in renewable-energy data centers using the Sonic Inference Engine, with real-time WebSocket connections for low-latency, scalable performance. It differentiates itself with a proprietary engine, dedicated hardware, built-in redundancy, autoscaling, GPU allocation, and access to over 188,000 models, enabling seamless integration without ML expertise. Its goal is to democratize access to generative AI for businesses by making it easy to add AI capabilities without heavy hardware costs or specialized know-how.

Company size
51-200
Company stage
Series A
Total funding
$66M
Headquarters
London, United Kingdom
Founded
2023

Recently Posted Jobs

Explore more opportunities at Runware

See more jobs