Infrastructure Ops Specialist

About Massed Compute

Massed Compute is building a modern GPU cloud platform for AI and high-performance compute workloads. Customers choose us when they need high-performance infrastructure with more flexibility, visibility, and support than they get from traditional options. Our work sits at the intersection of AI infrastructure, product design, systems thinking, and customer obsession.

Massed Compute is looking for an Infrastructure Operations Specialist to make sure every GPU we buy ends up in the right place, in service, and earning — and that we always know where all of it is

You’ll be the connective tissue between procurement, data center leasing, site operations, and solutions engineering. Those four functions each hold part of the picture: what we’ve bought, where we have space and power, what is actually racked and running, and what we’ve promised customers. That means deciding where new equipment goes across our site portfolio, tracking every shipment from purchase order to production, running deployments and redeployments, driving warranty and RMA work to resolution, and building the process that tells us what state our infrastructure is in now and what it will be in ninety days.

This role involves extensive logistics and coordination that will directly drive utilization and operational efficiency, and transparency that informs decisions and prioritization for multiple teams across the company. Additionally, you’ll define what has to be tracked and how, and partner with engineering and use coding agents to build out the tooling behind the monitoring and analytics.

This is a builder role. We’re looking for someone who can both run today’s deployments personally and build the tracking, process, and vendor discipline that let us run ten times as many without ten times the people.

In your first 8–12 months, you will have built a single, trusted view of what hardware we own and what state it is in, cut the time between equipment arriving and equipment earning, established a warranty and RMA process that resolves failures without anyone chasing them, and given procurement, capacity, and the commercial team a forward view of the fleet they can plan against.

Critically, you will have replaced a picture that lives in a handful of people’s heads and a dozen spreadsheets with a system that stays true on its own — so that "where is it, what state is it in, and when is it available" stops being a question anyone has to ask around to answer.

• Decide where equipment goes. Own placement across our site portfolio, balancing power and cooling headroom, rack density, customer adjacency, deployment speed, and what each site should be used for over the next year.

• Track everything from PO to production. Maintain the chain of custody for every shipment — ordered, in transit, received, staged, racked, cabled, burned in, accepted — and know at any moment what is where and what is late.

• Run deployments and redeployments. Coordinate the sequence across facilities, networking, storage, and platform teams, and move capacity between sites when demand or economics change.

• Own warranty, RMA, and servicing. Drive hardware failures through vendor and depot processes to resolution, track what we are owed, and hold vendors to their commitments.

• Own the fleet system of record. Select, implement, and maintain the asset and inventory system that tells us what we own, where it is, and what state it is in — including the reconciliation discipline that keeps it accurate

• Build the forward view. Design the reporting that shows what is landing when, what capacity frees up when, and where we are at risk so leadership and the commercial team can plan against reality rather than…