Head of Infrastructure Operations

Runware

3d•Remote

About The Position

As Head of Infrastructure Operations, you will own Runware’s physical and operational infrastructure: servers, storage and networking across every site, along with the team responsible for operating them. You will scale the company from a handful of pods to a global fleet and determine how the fleet is built, provisioned, monitored, maintained and expanded. This is a hands-on leadership role that combines setting the operating model and hiring the team with direct engineering involvement in terminals, racks and incident response. You will build and run infrastructure operations across pod sites in the US and Europe, covering site deployment through steady-state operations and decommissioning. You will own the server, storage and network layers end to end, establish repeatable and increasingly automated site deployment practices, and lead reliability, capacity planning, monitoring and incident management. The role also includes vendor and partner management, collaboration with hardware and platform engineering, using AI tools to improve operational leverage, and ownership of physical infrastructure security and compliance. The position reports to Risto Haukioja, Head of AI Infrastructure, and works closely with hardware, platform and inference engineering teams, as well as data center, power and hardware partners. Runware is remote-first, with in-person company meetings twice a year. Core collaboration hours are maintained, with flexibility outside those hours. The company describes its release cycles as fast and intense, followed by downtime.

Requirements

15+ years of experience in infrastructure engineering and operations, including several years leading the function at a company operating its own hardware at scale.
Track record of building infrastructure operations across multiple sites, regions or countries, including opening data centers, pods or colocation footprints.
Deep, current, hands-on Linux expertise, including kernel and driver behavior, performance tuning, bare-metal provisioning, configuration management and fleet automation.
Strong storage fundamentals, including NVMe, distributed and parallel file systems, object storage, and designing for the throughput and IOPS required for model loading and caching.
Strong networking fundamentals, including L2/L3 design, BGP, data center fabrics, high-bandwidth interconnects and network troubleshooting.
Experience hiring, leading and retaining strong engineers, setting clear standards, and developing sustainable on-call and incident practices.
Daily use of AI to accelerate work such as scripting, runbooks, analysis, documentation and debugging, and a clear perspective on how AI tools should change the operation of a small infrastructure team.
Ability to own outcomes in a fast-moving, ambiguous environment, make decisions with incomplete information, and be accountable for uptime, cost and delivery.
Ability to travel to sites in the US and Europe as needed.

Nice To Haves

Experience operating large GPU fleets for AI training or inference, including NVIDIA driver, Fabric Manager and NCCL stacks, and GPU health and failure management at scale.
Hands-on experience with RDMA fabrics such as InfiniBand or RoCE, 400G+ Ethernet, and related monitoring tools.
Experience with liquid-cooled, high-density, modular or containerized data center designs and site-level power and cooling integration.
Familiarity with the hardware supply chain, including server ODMs, component lead times, inventory and spares strategy, and RMA economics.
Strong automation stack background, including Ansible, Terraform, PXE and image pipelines, IPMI and Redfish, and observability tools such as Prometheus and Grafana.
Experience with Kubernetes or Slurm on bare metal.
Experience building or operating infrastructure to meet data residency, sovereignty or compliance requirements across jurisdictions.

Responsibilities

Build and run global infrastructure operations across pod sites in the US and Europe, from site bring-up, rack and stack, cabling, and power and cooling integration through burn-in, acceptance, steady-state operations and decommissioning.
Own the server, storage and network layers of the fleet, including Linux fleet management, bare-metal provisioning and imaging, firmware and driver lifecycle, high-throughput storage, and data center and inter-site networks.
Define and improve a repeatable, documented and largely automated playbook for bringing new sites online.
Own capacity planning, monitoring, alerting, on-call, incident response and post-incident reviews, and improve GPU, node and network reliability over time.
Hire, lead and develop data center, systems and network engineers across time zones, and establish operating standards for the team.
Manage colocation, power, connectivity and hardware vendors and partners, including capacity contracts, lead times, spares, RMA flows and site-level budgets.
Partner with hardware and platform engineering to ensure designs are operable at scale and feed field learnings into future servers and pods.
Use AI tools and LLM-based agents to automate runbooks, triage and tooling, and encourage the team to do the same.
Own operational security and compliance for physical infrastructure, including access control, hardening, asset tracking, audit readiness and data residency requirements.