About The Role
Volta builds and operates large scale GPU compute infrastructure for frontier AI customers. Every deployment is a project: a site, a customer, a hardware generation, a power and cooling envelope, a network fabric, and a date the cluster has to be producing useful work.
Our project teams own those deployments end to end. Each is roughly five senior engineers spanning data center operations, network engineering, and systems and platform engineering. They join during the discovery phase of a deal, when the technical shape and the commitments are still being formed, and stay with the project until the cluster is handed over to steady state operations.
As Head of Project Engineering you own that function. You lead multiple project teams delivering concurrently across sites, you are accountable for what Volta commits to technically during deal discovery and for whether we then deliver it, and you own the standard by which a cluster is declared done.
This is a leadership role with deep technical substance. You will not be running these deployments from a spreadsheet.
What You Will Be Doing
Deal Discovery and Commitments
Own technical input into deals from the discovery phase: site feasibility, power and cooling envelope, hardware and fabric options, deployment sequencing, and what is realistic on what timeline.
Pressure test what is being committed before it is signed. You are the person who says a date is not achievable while that is still cheap to hear.
Contribute the technical basis for cost models, bills of materials, and capacity assumptions, and make the assumptions explicit so they can be revisited when reality moves.
Support customer technical due diligence and, where relevant, lender or investor technical review.
Project Delivery
Hold accountability for delivery across a portfolio of concurrent deployments: scope, technical quality, and readiness against the committed date.
Establish the delivery model each project team follows, from facility handover through rack and stack, power on, fabric build, and burn-in to acceptance.
Own escalation. When a project is off track, you are the one who diagnoses whether it is a supply problem, a facility problem, a design problem, or a team problem, and who resolves it.
Allocate people across projects and manage the reality that senior specialists are scarce and demand is lumpy.
Own risk across the portfolio, including the dependencies that sit outside Volta.
Handover to Steady State Operations
Define and enforce what "done" means: acceptance criteria, performance validation, documentation, runbook completeness, monitoring coverage, and operational readiness.
Own the handover to steady state operations as a controlled transition with an agreed definition of complete, not a gradual disengagement.
Make sure the operational and SLA commitments Volta has made are actually supportable by what the project team leaves behind.
Close the loop back into engineering: what went wrong, what was manual that should not have been, and what needs to change in the platform or the process before the next site.
Organisation and People
Build and lead the project engineering function: team leads, project engineers, and the senior specialists who make up the teams.
Hire senior technical people into a role that requires travel, ambiguity, and site presence, and be honest with candidates about what that involves.
Run the management basics properly: 1:1s, performance conversations, development, and succession for a function where key person risk is real.
Establish a career path for project engineers so that the role is a destination rather than a rotation people try to leave.
Partners and Vendors
Own the technical relationship with OEMs and integration partners through delivery, including holding them to their commitments.
Work with facility and colocation partners on demarcation, handover, and the boundaries of who owns what physically and operationally.
Manage partner-delivered scope where deployments use external field services or managed services capacity.
Standards and Repeatability
Turn deployments from bespoke projects into a repeatable process. The second site should be materially easier than the first, and the fifth should not need you in the room.
Own the reference deployment model, standard acceptance test suites, and the documentation set every project produces.
Work with platform and network engineering so that what the project teams do by hand today becomes automated tomorrow.
What You Bring
8+ years in data center, infrastructure, or large scale technical delivery, with at least 3 years leading multiple teams or leading through team leads.
Track record delivering large scale compute deployments end to end, at a scale where power, cooling, and logistics are constraints rather than details.
Real technical depth in at least two of: data center facilities and power and cooling, high performance network fabric, or compute and systems engineering. Enough credibility in the others to lead senior specialists who know more than you do about their own area.
Familiarity with GPU cluster deployment: rack scale systems, high density power, liquid cooling, and the acceptance and validation work that proves a cluster is fit for training workloads.
Experience owning delivery against commercial commitments, where dates and acceptance criteria have contractual and financial consequences.
Multi-vendor OEM and partner management under delivery pressure, including holding vendors to schedule and quality without breaking the relationship.
Experience running concurrent projects with shared, constrained specialist resource.
Comfortable in commercial conversations: contributing to deal shaping, feasibility, and cost basis, and translating technical constraints for commercial and customer audiences.
Willing and able to travel to sites and partner locations regularly, including during critical delivery phases.
Clear communicator who can hold an uncomfortable position with senior stakeholders and customers when the technical reality requires it.
Nice to Have (But Not Essential)
None of these are required.
Experience with NVIDIA reference architectures and rack scale systems, and with the certification and validation process that comes with them.
Direct liquid cooling deployment experience, including CDU integration and facility demarcation.
Background at a neocloud, hyperscaler, HPC centre, or specialist data center integrator.
Experience deploying in the Nordics or in sites with unusual power, climate, or grid characteristics.
Exposure to technical due diligence for lenders or investors on infrastructure projects.
Familiarity with structured delivery frameworks, and a clear view on where they add value and where they become overhead.
Experience introducing automation or tooling into a delivery function that previously ran on people and spreadsheets.
Fluency with AI-assisted workflows for planning, documentation, and reporting.