SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Lumafield is a manufacturing technology company founded in 2019, headquartered in San Francisco with an office in Cambridge, MA. The company provides industrial CT scanning and AI-driven tools that give engineers unprecedented visibility into product manufacturing and inspection.
As an Infrastructure Engineer focused on On-Prem Systems, you will own the design and deployment of Voyager, Lumafield's CT data platform, for on-premises installations. This role is critical for defense and air-gapped programs that cannot send scan data to the cloud, and for production lines running high-speed scanners that cannot tolerate cloud upload latency.
You will be responsible for the complete lifecycle of on-prem solutions: designing standalone, high-performance systems including compute nodes, storage topology, networking, power, and rack layout. You will prototype and profile systems to benchmark throughput, network bandwidth, and memory contention. A key responsibility is creating engineering deliverables that Manufacturing can execute independently: hardware and software BOMs, wiring and rack diagrams, network and IP plans, and provisioning and test work instructions.
You will harden appliances for customer deployment, implementing encryption, network segmentation, firewall policies, and update mechanisms that function in air-gapped environments. You will run Kubernetes on fixed capacity, managing node pools, worker placement, and underlying storage and database services. You will partner with Manufacturing, Supply Chain, and Field Services to transition prototypes into repeatable products, maintaining accountability for field deployments.
This is a hands-on, in-person role based in San Francisco. You will feed on-prem constraints back into the technology roadmap, ensuring the next generation of the pipeline aligns with the hardware it will run on.
REQUIREMENTS:
- 3+ years building and operating physical server infrastructure (racking, cabling, power sizing, post-deployment debugging)
- Linux server administration, network troubleshooting, Docker deployments, and system productionization
- Linux system configuration, best practices, and workflow automation with Ansible and scripting
- Hands-on bare-metal provisioning experience: PXE/netboot, cloud-init, golden images, firmware/BIOS configuration via Redfish or IPMI
- Real networking depth: VLANs, subnetting, DHCP, DNS, firewall policy, link aggregation
- Working knowledge of Kubernetes on fixed capacity
BONUS:
- Production experience with ZFS (pool topology, resilver behavior, native encryption)
- Experience shipping hardware and manufacturing handoff, ECO processes, or field service of deployed systems
- Familiarity with NVIDIA GPU compute on ARM platforms (DGX, Jetson, CUDA) or air-gapped/sensitive data deployments
