Senior Network Engineer

Overview

Join our team to design, automate, and operate advanced datacenter and hardware validation lab networks supporting global accelerator and HPC cluster projects. You will drive deep L2/L3 network architecture, automation, and operational excellence for environments critical to hardware bring-up, precision timing, and provisioning at scale.

Responsibilities

• Architect, deploy, and maintain complex datacenter and lab fabric networks, focusing on reliability, scalability, security, and proactive cost management through lifecycle decommissioning.
• Design and operate L2/L3 solutions: VLANs, loop prevention, LAG, IPv4/IPv6 planning, SLAAC, DHCP relay/helpers, PXE boot flows, and seamless BGP integrations (including eBGP peering, prefix filtering, uplink preferences, per-rack/site route origination).
• Engineer datacenter fabrics: leaf/spine topologies, EVPN/VXLAN overlays, L2-to-L3 migration, and IPv6-first operations (first-hop security/RA guard, /64 per rack scale, DHCPv6 relay, and MAC-stable configs).
• Manage and optimize timing infrastructure: PTP (IEEE 1588) deployments, grandmaster and clock placement, sync quality, and troubleshooting against strict latency/timing budgets.
• Operate and validate optical/physical plant: fiber/connector types, optical switching, traffic generators, optical signal validation, cable/fiber characterization with hardware timestamping.
• Enforce and enhance access security: port ACLs, 802.1X/MAB, RADIUS, device onboarding, quarantine/remediation VLANs.
• Lead automation and network software: Python for device APIs and operator tooling, Jinja2 templating for reusable config, distributed version control/config as code, CI review gates, app-level intent modeling, and source of truth system integration.
• Maintain clean operational practices: data hygiene on inventory/roles/ports/serials, drift detection, reconciliation automation, thorough documentation.
• Drive operational delivery: incident response/oncall support, break/fix prioritization, method of procedure authorship and peer review, blast radius control, pilot-to-scale rollout, and vendor/circuit management.
• Collaborate cross-functionally with platform, hardware bring-up, provisioning, and lab operations teams for shared results beyond the planned scope.

Minimum Qualifications

• 10+ years of hands-on network engineering, datacenter/lab operations, or validation network support in large-scale or HPC environments.
• Advanced proficiency with at least one modern datacenter NOS (Arista EOS, Cisco NX-OS, Juniper, or SONiC) and the ability to adapt to others, including automated/API-driven operation and rollback.
• Mastery of L2/L3 protocols: VLANs, loop prevention, link aggregation (LAG), IPv4/IPv6, DHCP relay, PXE boot, SLAAC, router advertisements, BGP (eBGP, prefix policy, uplink preference), EVPN/VXLAN, and multi-site leaf/spine fabric design.
• Demonstrated experience with precision timing (PTP/IEEE 1588, grandmaster/boundary/transparent clocks), and lossless Ethernet (PFC) in AI/HPC environments.
• In-depth understanding of fiber optics, optical switching, connector standards, and commercial test/validation tooling.
• Proven skill in Python network automation, device APIs, Jinja2 templating, distributed version control processes (branch/review/land/revert), config as code, and declarative intent modeling.
• Strong Linux networking fundamentals, service/process management, cookbook config management (such as Chef), and network observability/capture tooling.

Preferred Qualifications

• Hands-on practice with intent-driven provisioning, integration of declarative source-of-truth systems, and advanced automation frameworks for network CI/CD and automated reconciliation.
• Experience operating and optimizing networks for AI/HPC clusters: scale-up vs scale-out topologies, RDMA-capable fabrics, tail-latency and collective communication troubleshooting.
• Expertise in large-scale fiber/optical plant management, including cable/fiber validation, traffic generators, and fiber plant metrics.
• Advanced observability: developing and managing time series metrics, topology visualization, and low-noise alerting tailored for lab/test environments.
• Practice in lifecycle operations: multi-site decommissioning, cost/circuit tracking to zero, chargeback validation, and vendor/circuit carrier negotiation.
• Vendor evaluation: feature validation, optics compatibility testing, and objective switch selection methodology.