SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Fuse Energy is a venture-backed energy startup ($200M+ raised from Balderton, Accel, Creandum, Lowercarbon, Ribbit and others) building an integrated energy company combining solar, batteries, grid infrastructure, AI, and distributed energy for consumers.
You will design, deploy, and operate the network fabric for Fuse Energy's multi-tenant AI cluster, handling high-performance compute and storage fabrics carrying RDMA traffic between GPUs, tenant-facing and management networks, firewalling, tenant isolation, and out-of-band infrastructure. You own the fabric from architecture through day-2 operations. Beyond the data centre, you will own the office network and serve as the networking authority for the company.
Key responsibilities include:
- Design and operate lossless, RDMA-capable fabrics (RoCEv2, InfiniBand) for GPU compute and storage, including QoS, congestion control, and buffer tuning at scale
- Build and manage leaf-spine data centre fabrics with routed underlay and overlay design (BGP, EVPN/VXLAN)
- Implement and maintain per-tenant network isolation across compute, storage, and management planes
- Automate network provisioning, configuration, and validation using infrastructure-as-code practices (Ansible, Python, NetBox), deployed through CI
- Build telemetry and observability for the fabric with flow-level and buffer-level visibility, dashboards, and alerting
- Troubleshoot performance issues end-to-end, from optics and cabling through switch buffers to NIC/DPU configuration and collective-communication behaviour
- Operate the out-of-band management network, console access, and remote recovery paths
- Support tenant onboarding with segmentation, addressing, bandwidth guarantees, and capacity planning
- Write clear design documentation capturing decisions, rationale, and rejected alternatives
- Own and maintain the office network including wired/wireless infrastructure, firewalling, VPN/remote access, and office-to-data-centre connectivity
- Upskill colleagues on networking through documentation, run-throughs, and pairing
Requirements:
- 5+ years as a network engineer operating production data centre networks
- Strong dynamic routing experience (BGP in particular), plus overlay/encapsulation design and troubleshooting (EVPN/VXLAN)
- Hands-on experience with leaf-spine/Clos fabric design and operation
- Experience with modern data centre network operating systems and comfort in the Linux networking stack, not just vendor CLI
- Practical RDMA fabric experience: lossless Ethernet (RoCEv2 with PFC/ECN/DCQCN tuning) or InfiniBand, understanding why lossless behaviour matters for GPU workloads
- Network automation as a working practice: scripting (Python), configuration management (Ansible), config generation from source of truth, version-controlled changes
- Solid Linux administration fundamentals: ability to debug from host side and switch side
- Experience with network telemetry and monitoring (Prometheus/Grafana, sFlow/IPFIX, streaming telemetry)
- Experience running corporate/campus networks: wired and wireless, switching, NAC/802.1X, VPN and remote access
- Clear communicator who enjoys teaching
Bonus experience: GPU cluster networking (NVIDIA Spectrum-X or Quantum InfiniBand, ConnectX/BlueField NICs and DPUs, UFM, SHARP), container networking (CNI, BGP integration), collective-communication libraries, multi-tenant VRF-based isolation, enterprise firewalls (FortiGate, Palo Alto), storage networking (NVMe-oF), bare-metal provisioning (MAAS, PXE, Redfish), optical layer at 200/400/800G, greenfield data centre network build, CCNP/CCIE or equivalent.
