Description
This role owns the journey of every machine in the fleet: from the moment a server is racked, cabled, and powered on, to the moment it is fully configured, validated, and available for users. Your mission is to make that journey zero-touch.
You will design and build the automation pipeline that takes a machine through discovery, firmware and BIOS configuration, OS installation, configuration management, health validation and burn-in, and finally handoff into production, treating each stage as code that lives in Git, runs through CI/CD, and can be reviewed, tested, and rolled back like any other software.
The guiding principle is GitOps for physical infrastructure: the desired state of the fleet is declared in a repository, and automation continuously reconciles reality against it. A new machine shows up as a commit; a decommission is a deletion; drift is detected and corrected by the pipeline, not by a person with a checklist.
Responsibilities:
- Design and build the end-to-end machine lifecycle pipeline: from power-on and network boot through OS install, configuration, validation, and production handoff.
- Automate hardware bring-up via out-of-band management (BMC, Redfish, IPMI): firmware updates, BIOS settings, boot order, and inventory discovery.
- Automate OS provisioning with network boot (PXE / UEFI HTTP boot) and unattended installation, so no one ever installs a machine by hand.
- Write and maintain the Ansible and Python that configure machines into their final roles, replacing manual runbooks with reviewed, versioned code.
- Apply GitOps and CI/CD principles to the fleet: desired state in Git, changes through merge requests, pipelines that test and apply them, and reconciliation that catches drift.
- Build automated validation and burn-in: health checks, stress tests, and acceptance criteria a machine must pass before users ever see it.
- Model the lifecycle as a state machine (new, provisioning, validating, in-service, needs-repair, decommissioned) with clear, automated transitions and an auditable history.
- Instrument the pipeline with metrics and logging so we always know where a machine is in its lifecycle, and where the process is slow or failing.
- Work with the HPC and datacenter teams to fold their hard-won operational knowledge into the automation, one stage at a time.
Qualifications:
- A smart, curious engineer who learns fast and is genuinely excited by the challenge of automating physical infrastructure end to end. This matters more to us than any specific line on your resume.
- Strong Python for building automation, tooling, and services, not just scripts.
- Hands-on Ansible experience: writing playbooks and roles you would be happy to code-review, not just run.
- A solid grasp of CI/CD principles: pipelines, testing, staged rollouts, and the discipline of driving change through version control.
- An automation-first, GitOps mindset: you believe infrastructure state belongs in Git, and that any task done by hand twice should be code.
- Working knowledge of Linux: comfortable with the boot process, system services, and debugging when a machine does not come up the way it should. Depth here is a real plus, but interest and trajectory count.
- Sound engineering judgment: you design workflows that fail safely, retry sensibly, and leave an audit trail.
- Clear communication and the patience to turn tribal operational knowledge into reliable, documented automation.
Nice to Have:
- Experience with bare-metal provisioning tooling such as MAAS, Tinkerbell, Foreman, Ironic, or a home-grown equivalent.
- Familiarity with out-of-band management: BMCs, Redfish, IPMI, and vendor variants like iDRAC or iLO.
- Exposure to hardware validation and burn-in: stress testing, firmware qualification, or failure prediction at fleet scale.
- Experience with GitOps tooling or declarative infrastructure management in general.
- Prior work in datacenter, HPC, or large-fleet environments where machines number in the hundreds or thousands.
Comp: 150-250K base + Bonus