Deployment Engineer
As a Deployment Engineer you own the last — and most critical — mile: turning a designed AD-X AI factory into a live, training-ready system on the customer's floor. You'll travel to sites worldwide, lead on-site commissioning, and hand over systems that run from day one.
You're the technical owner on the ground, coordinating facilities, electrical, networking, and customer teams to bring power, cooling, fabric, and orchestration online together.
What you'll do
Lead on-site installation and commissioning of AD-8 servers and full AD-X deployments — power, direct-liquid cooling, networking, and orchestration.
Run acceptance testing and burn-in, validating thermal, power, and fabric performance against spec before handover.
Integrate systems with the customer's environment — schedulers, storage, identity, and monitoring.
Coordinate with facilities, electrical, and customer teams to keep deployments on schedule and safe.
Write the runbooks and train customer operators so handover is clean and the system is theirs to run.
Feed field learnings back to our hardware and systems teams to make the next build better.
What we're looking for
4+ years deploying or operating data-centre, HPC, or GPU infrastructure in the field.
Hands-on with rack-and-stack, structured cabling, high-speed networking (InfiniBand / Ethernet fabric), and power distribution.
Working knowledge of liquid cooling (direct-to-chip or immersion) and data-centre facility systems.
Calm, methodical, and comfortable being the on-site technical owner under deployment pressure.
Willing to travel internationally up to ~50%, sometimes at short notice.
Nice to have
Experience commissioning GPU clusters for AI training at scale (100s–1,000s of GPUs).
Familiarity with Slurm or Kubernetes, BMC/IPMI, and cluster monitoring stacks.
An electrical or mechanical engineering background.