Network Architect

Runware

2d•Remote

About The Position

Runware is hiring a senior, hands-on Network Architect to own the design, construction, and operation of its end-to-end network. This includes GPU cluster fabrics in data centers, the backbone and edge that deliver inference results to customers, and cloud and hybrid connectivity linking partner capacity to the platform. Network performance directly affects product latency and cost per request, and this role will establish clear ownership of that domain. The architect will set network strategy and the multi-year roadmap, build systems directly, define operating and capacity-planning practices, and lead through design authority across Infrastructure, Platform, and ML Engineering. The role begins without direct reports but includes setting standards, mentoring engineers, and defining the hiring plan and operating model as the network footprint grows. Runware expects AI tooling to be used as a core part of the work. The position is remote-first, with in-person gatherings twice a year and core collaboration hours. Hours outside those blocks are flexible. Release periods can be intense, followed by expected downtime.

Requirements

15+ years of experience in network engineering and architecture, including at least 5 years as design authority for large-scale production networks at a cloud provider, hyperscaler, CDN, HPC/AI infrastructure company, or carrier.
Hands-on experience building data center, backbone, and cloud networks from the ground up.
Current expertise in spine-leaf Clos fabrics, EVPN/VXLAN, BGP at scale, 400G/800G Ethernet, RoCEv2 and lossless Ethernet (PFC, ECN, DCQCN), and InfiniBand for GPU clusters.
Strong backbone and edge experience with multi-region interconnect, transit and peering, anycast, traffic engineering, and DDoS mitigation, including a record of measurably reducing customer-facing latency.
Cloud and hybrid networking experience across at least two hyperscalers and bare-metal or colocation environments, including VPC design, private interconnect, and VPN at scale.
Experience designing and operating network operations practices covering observability, change management, incident response, and on-call.
Proven capacity-planning experience, including forecasting models and associated bandwidth, port, interconnect, and procurement plans.
Practical network automation experience with Python or Go, infrastructure as code for network configuration, CI-style validation, and streaming telemetry.
Uses AI tooling to accelerate network design, configuration and automation development, telemetry analysis, and documentation.
Experience leading high-performance engineering teams, including hiring, setting standards, and delivering under pressure; able to lead through influence without direct reports.
Strong ownership and ability to create clarity and drive work to completion in a fast-moving, ambiguous environment.
Based in or able to maintain full working overlap with US Pacific time.

Nice To Haves

Experience networking large GPU clusters for AI inference or training, including NVIDIA Spectrum-X, Quantum InfiniBand, NCCL tuning, or GPUDirect RDMA, and production-data-backed views on Ethernet versus InfiniBand.
Hands-on experience with network operating systems and disaggregated hardware such as SONiC, Cumulus, Arista EOS, Junos, or NVIDIA Cumulus/Spectrum platforms.
Experience with Ultra Ethernet, 800G and 1.6T optics, co-packaged optics, or other emerging fabric technologies.
Experience operating peering and transit for a latency-sensitive product, including ASN management, route optimization, and IXP presence.
Experience with eBPF, DPDK, SmartNICs/DPUs, or kernel-bypass networking on the host side.
Prior experience at a startup or scale-up building the network function from zero.
Public contributions such as RFCs, open-source network tooling, conference talks, or published post-mortems.

Responsibilities

Own end-to-end network architecture and define the target architecture and multi-year roadmap for data center fabrics, backbone and edge, and cloud and hybrid connectivity.
Design and build GPU cluster fabrics for multi-node inference and training, including topology, high-speed Ethernet, RoCEv2, InfiniBand, lossless transport, and congestion control.
Own backbone and edge architecture, including multi-region interconnect, transit and peering, anycast, traffic steering, DDoS and edge security, and customer response paths.
Design VPC, VPN, and private interconnect across hyperscalers, bare-metal providers, and colocation sites.
Establish network operations practices for monitoring, telemetry, observability, change management, incident runbooks, and on-call; build missing capabilities.
Forecast bandwidth, port, and interconnect needs and translate forecasts into procurement and network build plans.
Automate network configuration and operations using infrastructure as code, intent-based provisioning, automated validation and rollback, and telemetry-driven remediation.
Use AI tools for design exploration, configuration generation, log and telemetry analysis, documentation, and automation code; help set usage standards for the infrastructure team.
Set network standards, review designs across Infrastructure, Platform, and ML Engineering, mentor engineers, and define the network engineering hiring plan and operating model.
Evaluate and select network hardware, optics, carriers, and interconnect partners, and negotiate on performance and cost.
Design, configure, and troubleshoot networks hands-on.