Head of Advanced Support

Description

Location: US (travel to customer sites and to R&D in Israel)

About DriveNets DriveNets is a leader in high-scale networking software for AI infrastructure and service providers. The company pioneered a disaggregated networking architecture that transforms the economics of large-scale networks while maximizing performance, utilization, and operational efficiency. DriveNets-powered networks are deployed by global leaders, including AT&T and Comcast, supporting more than 30% of total U.S. internet traffic. Its AI group applies the same approach to GPU clusters, delivering high-performance Ethernet fabrics on Broadcom Tomahawk-class switches. With over $1B raised, DriveNets continues to push the boundaries of modern networking infrastructure. DriveNets and AMD are now taking the next step together. Neoclouds that want AMD capacity can buy a complete AI Factory from DriveNets, built on AMD’s Helios rack-scale platform. DriveNets is accountable for the whole thing: every network in the data center (front-end, back-end scale-out and the connection to the in-rack scale-up network), cluster bring-up and installation, the operating software, and multi-year maintenance and support (M&S) across the entire estate, including the GPU compute trays. Our premise is simple: AMD + DriveNets ≥ NVIDIA. NVIDIA sells a vertically integrated stack. We compete as the open, multi-vendor alternative, and AMD silicon plus a DriveNets fabric, software layer and service organization has to give customers the same confidence. The hard part is operations, not specs:

• Helios is brand new. It is a first-generation rack-scale platform with no installed base, no field history and no library of known failures. We have to build that knowledge ourselves, together with AMD, while the first customers go live.
• Neoclouds expect hyperscale uptime. Their customers are AI labs and inference companies that measure us by availability, time to first token and how quickly a degraded node is found and fixed. Every idle GPU costs them money.
• We are accountable for components we did not build. GPUs, compute trays and scale-up switches come from AMD and its manufacturing partners. When something breaks, the customer calls DriveNets. We triage, coordinate the vendors and own the outcome.
• Hardware faults are expensive. A failed GPU means swapping a whole compute tray. Every no-fault-found return ties up spares and leaves the real problem in the customer’s cluster, so finding the correct root cause quickly is critical.
• Deployments are on-prem and often air-gapped. Data reaches us only with customer approval, so tooling, diagnostics and processes have to work under that constraint.

Our answer is the AI Cluster Orchestrator, DriveNets’ operations platform for the AI Factory. Foundation covers inventory, topology, day-zero validation and lifecycle. Performance & Reliability detects, triages and pinpoints faults and is backed by a shared Diagnostics library of failure modes. Care handles support cases, phone-home and evidence collection with customer approval. Security covers estate security and drift. It is meant to turn monitoring data into action, and the people in these roles will shape it as its first and most demanding users. The first production AI Factory deployments, where DriveNets owns the whole stack, are planned for early 2027. We are building the support and customer-facing organization now, from scratch, to be ready for them. The Role We are looking for a hands-on leader to build DriveNets’ advanced support organization for the AI Factory. Reporting to the CTO, AI group, you will own how hard problems get solved for our neocloud customers, from the first Helios deployment onward. Your organization sits between the customer, our own R&D and AMD and its partners. You will hire the team, define the playbooks and set the standard for what support from DriveNets means. For the first year this is a build-and-do role: you will often be the most senior engineer on a critical case. As DriveNets expands you will lead Tier 3 support engineers. Initial team: three T3 engineers, growing with the customer base

Roles and Responsibilities:

Build the organization

• Hire and lead T3 engineers (compute/GPU, fabric, software stack), and define the profiles, specializations and on-call model.
• Design the support operating model end to end: severity definitions, SLAs, escalation paths from T2 to T3 to T4 (DriveNets R&D, AMD, hardware partners), and handoffs with case managers and TAMs.
• In year one, lead a team that covers T2-level work as well while Helios knowledge is being built, and plan the move to a mature model where most issues are resolved at the first line through ACO.

Own the hard cases

• Be the escalation owner for critical and multi-vendor incidents: run war rooms, coordinate AMD and partner engineering, and own customer communication until resolution.
• Run post-incident reviews and make sure fixes reach product, firmware, service packs and the Diagnostics library.
• Work with logistics on RMA decisions, spares positioning and the vendor back-to-back service agreements, with a strong focus on reducing no-fault-found returns.

Shape the product and the knowledge base

• Act as the main internal customer of the Care and Performance & Reliability: define requirements, validate workflows and turn field reality into roadmap input.
• Own the field-facing Diagnostics process: make sure every new failure signature is documented, referenced and validated (with CRE) before it ships to customers, and track how often each one occurs across the fleet.
• Build a working technical relationship with AMD’s Helios support and engineering teams, including joint training, shared diagnostics and aggregated field feedback that helps AMD improve hardware and software.

Customers and accounts

• Through the TAMs, give every AI Factory customer a trusted technical owner who knows their sites, clusters and history.
• Join customer onboarding and design-partner reviews, and represent support in pre-sales discussions on M&S scope and SLAs.
• Report service health to leadership: CSAT, resolution time per severity, mean time to detect, pinpoint and resolve, NFF rate and repeat-issue trends.

What success looks like in the first year

• The first AI Factory customers go live with a staffed, trained and on-call advanced support team and documented escalation paths to AMD and partners.
• Critical incidents have one clear owner and are resolved within SLA, and customers name DriveNets support as a reason to expand.
• NFF and mean time to pinpoint trend down quarter over quarter, driven by ACO and a growing, validated Diagnostics library.
• A clear, measured plan exists to shift the issue mix toward first-line resolution as the platform and knowledge mature.

Requirements

Qualifications

Required

• 12+ years in technical support, escalation engineering or field engineering for data-center infrastructure, including 5+ years leading senior support or escalation teams.
• Hands-on depth in at least one of GPU servers and accelerators, HPC/AI clusters, or data-center networking (Ethernet/RDMA, RoCE, BGP/EVPN), and enough fluency in the others to lead experts across all three.
• Track record of running multi-vendor incidents where you owned the customer outcome without owning every component.
• Experience setting up or scaling a support function: tiers, SLAs, on-call, knowledge management, KPIs.
• Strong RMA and hardware-lifecycle experience (failure analysis, spares strategy, NFF reduction).
• Excellent written and verbal communication, and able to stay calm and credible with a customer’s executives during an outage.
• Willing to travel to customer sites and to R&D in Israel.

Nice to have

• Support or field leadership at a GPU/AI infrastructure company (e.g. NVIDIA, AMD, Dell, HPE, Supermicro), a networking vendor, or a neocloud/GPU cloud provider.
• Familiarity with AMD Instinct, ROCm/RCCL, or rack-scale AI systems.
• Experience with on-prem or air-gapped environments and support tooling (ServiceNow, Jira, Zendesk).
• Has used AI agents and automation to scale a support organization.

Why join:

This is a ground-floor role in a new business. You will help define how AI factories built on AMD are operated and supported, work with AMD’s engineering teams on a brand-new platform, and have direct influence on a product (ACO) that is designed around the work you do. More About DriveNets Based in Israel with locations in Romania, US, India and Japan as well as extended teams, DriveNets operations cover more than twelve countries. With recognition by industry analysts and through partnerships with market leaders such as AMD, Broadcom, Dell and others, DriveNets is pushing market momentum, delivering the scale and efficiency that modern AI workloads demand. Visit our website: https://drivenets.com/company/