About the RoleWhat You'll Do- Own the end-to-end network architecture for Distributed Compute networks: site connectivity, backhaul, the fleet-wide overlay, and the interconnect within each node. Standardize the per-site network stack (routing, switching, firewall, out-of-band management) so new sites come online quickly.
- Lead fiber and carrier strategy for the fleet. Evaluate dark fiber vs. lit services, and build vs. lease decisions per site cluster. Select topologies that scale with site density, and manage carrier/ISP selection, circuit ordering, and failover across fiber, fixed wireless, and cellular. You'll build on existing work with Base's fiber consultants and take over this function as it scales.
- Build and operate the fleet overlay and its automation. Establish secure site-to-site and site-to-cloud connectivity, BGP routing, and traffic engineering for inference workloads. Automate configuration generation, zero-touch provisioning, and monitoring so network state is managed in code.
- Instrument, secure, and support the network in practice. Track latency, throughput, packet loss, and circuit health in Base's telemetry and Grafana stack. Define segmentation between compute, battery control, and management planes, along with certificate management and audit-ready access controls. Support the first installs and site bring-ups in the field, then reduce on-site involvement as processes mature.
What You'll Bring- 5+ years running production networks
- Deep, hands-on routing and switching: BGP, OSPF, VLANs, VRFs.
- Experience with site-to-site connectivity at scale: SD-WAN or overlay networks (WireGuard, IPsec) across tens to hundreds of locations.
- Familiarity with fiber networks and topologies
- Experience procuring and managing circuits with carriers
- Comfort in the field. Some of this job happens at a site in a utility territory
About the TeamDistributed Compute is Base's program for deploying modular datacenters on top of our battery fleet, turning distributed energy sites into a distributed inference network. That scale is a networking problem as much as a power problem. Hundreds of small sites, each running GPU compute, spread across utility territories, and every one of them needs reliable, secure, high-throughput connectivity, with the fleet behaving like one coherent network. This is the first dedicated network engineering hire on the program, so the architecture decisions are still open and the constraints are real: residential and utility sites, power budgets, physical access. What you build will carry production inference traffic within months, not years.
Please note: Base is a startup, which means priorities shift and evolve quickly. Your role may expand or change based on the needs of the business at any given time, so the responsibilities listed may not be exhaustive.
