Job Opportunities API

1. ABOUT THE ROLE The global marketplace for bandwidth. Share aggregates telco infrastructure into a single intelligence layer, giving ISPs, hyperscalers, and AI companies a seamless path to deliver connectivity into every home and business. We're hiring a VP of Network Engineering to lead the function that makes this real. This is the senior-most networking role at Share, reporting directly to the founders. You will own the entire technical estate (backbone, aggregation, access, peering, the systems they depend on, and the partner integrations they touch) and the team that builds and runs it. This is a role for someone who wants to create, not just operate. The mandate is to design the protocols, define the standards, and develop the reference architectures that an entire ecosystem of partner ISPs will inherit from us, and to leverage AI across the operational surface so the network behaves more like a system that thinks than a stack of boxes that's been configured. The expectation is still hands-on. You'll set technical direction, run a small high-leverage engineering team, and stay close enough to the metal that you can sit on a BGP debug call at 2am and not be the bottleneck. This is also a deeply partner-facing role: you'll absorb partner ISPs into our platform and push each one past whatever ceiling they walked in with. 2. WHAT YOU'LL OWN The network as a system - Design the protocols, standards, and reference architectures that the rest of the ecosystem inherits from us. Partner ISPs across the continent will adopt the patterns you set (peering policy, hand-off contracts, service profiles, security posture) because conforming to Share's baseline is how they unlock the speeds we deliver. This is closer to standards work than to operations management: it ships as configuration, but it lives as doctrine. - Make the architectural calls that don't have a textbook answer yet: how a shared physical backbone is sliced across many partner ISPs at multi-tenant scale, how policy is expressed once and enforced everywhere, where the control plane goes next, what we retire, and what we keep. - Treat the network as a system that learns. Drive the use of AI in operations: anomaly detection across flow and streaming telemetry, predictive failure analysis on optical and routing layers, automated triage and runbook execution, traffic-engineering and peering decisions informed by models, and config-drift detection across partner hand-offs. You won't build the models from scratch, but you'll set the bar for where AI replaces human reaction time and where it doesn't. The routing and peering core - Own the strategy for our BGP edge: IP transit with Workonline, WIOCC, MTN/Bayobab, Afr-IX, and Skytrend, plus public and private peering at KIXP and LINX, and PNIs with the hyperscalers. - Own the policy posture: filter chains, route policy, communities, RPKI, and traffic-engineering logic across our border fabric and downstream. - Direct MPLS in the backbone (L3VPN and EVPN for partner isolation, services delivery, and multi-tenant scale) and decide where the control plane goes next (LDP vs. SR, address-family strategy). - Negotiate and establish new peering arrangements (content, CDN on-nets, cache fills, hyperscaler PNIs) and tune the mix for cost, performance, and resilience. Aggregation, access, and the services plane - Own the disaggregated, white-box switching strategy: modern NOS stacks on commodity silicon (SONiC, IP Infusion OcNOS, and similar). You should have a position on vendor lock-in, and you should have already acted on it elsewhere. - Direct L2/L3 aggregation (VLANs, MC-LAG/EVPN-VXLAN, QoS, clean L2 hand-offs to partner ISPs) and BNG strategy (PPPoE/IPoE termination, QoS, CGNAT where needed, CoA-driven session control that hooks into our AAA layer). - Own passive optical access (GPON, XGS-PON, OLT/ONU strategy, service profile design) and the fiber plant itself: ODF and splice planning, OTDR traces, lo