Senior AI Infrastructure Engineer, Physical Infrastructure
About this role
Scale bottlenecks and hardware instability in large GPU training clusters impede machine learning research and model deployment across defense technology programs. Anduril Industries addresses this challenge by hiring a Senior AI Infrastructure Engineer to lead the vision, execution, and long-term stability of GPU training operations. The CorpTech Infrastructure Engineering team builds foundational infrastructure that allows internal engineering and research teams to deploy scalable systems without becoming infrastructure experts.
This practitioner will take absolute ownership of cluster robustness, ensuring high-performance GPU systems remain fault-tolerant and highly available for company-wide ML initiatives. Responsibilities include building automated self-healing mechanisms for hardware faults, tuning NCCL and high-speed networking, and optimizing Kubernetes and Ray scheduling. Ideal candidates possess extensive hands-on datacenter experience supporting massive GPU fleets and thrive on replacing manual triage with deep observability and automation.
What you will do
- Rack, stack, cable, and bring up physical GPU compute hardware including power, cooling, firmware, and burn-in validation.
- Build and tune high-performance interconnect fabrics such as NVLink, InfiniBand, and RoCE connecting large GPU clusters.
- Integrate high-performance parallel storage systems like VAST, DDN, and Weka to sustain multi-modal dataset throughput.
- Automate cluster deployment, configuration, and infrastructure-as-code pipelines for firmware, drivers, and fabric management.
- Operate and extend Kubernetes and Run:AI environments for GPU scheduling, quota management, and multi-tenant isolation.
- Own fleet health through continuous monitoring, alerting, and rapid triage of hardware and network errors.
- Act as the technical escalation point to debug workloads and partner with internal teams on emerging compute requirements.
What you need
- Ten or more years of hands-on experience in infrastructure, HPC, or datacenter engineering supporting GPU compute at scale.
- Direct experience bringing up and managing modern GPU systems such as H200, B200, and B300 hardware.
- Familiarity with high-performance interconnects including NVLink, InfiniBand, RoCE, and Spectrum-X in large clusters.
- Experience implementing high-performance parallel storage solutions like VAST, DDN, Weka, or Lustre.
- Demonstrated proficiency with Kubernetes and GPU scheduling tools like Run:AI.
- Strong automation background focused on repeatable deployment pipelines rather than manual processes.
- Physical capability to lift and move fifty or more pounds while performing datacenter hardware tasks.
- Eligibility to obtain and maintain an active U.S. Top Secret security clearance.
Nice to have
- Experience with NVIDIA NVL72 rack scale systems
- Experience supporting LLM token serving and inference infrastructure alongside training clusters
- Network fabric tuning experience for RoCE and InfiniBand at scale
- Familiarity with GPU and network observability tooling such as DCGM and fabric telemetry
- Experience supporting infrastructure as a shared platform serving multiple internal customer teams
Benefits
- Top-tier benefits for full-time employees supporting health and recovery
- Competitive equity grants included in the majority of full-time offers
Skills and tech stack
Questions candidates ask
What is the salary range for this role?
The salary range for this position is $166,000 to $220,000 USD. Actual salary offers may vary based on work experience, education, critical skills, and business considerations.
What are the physical demands of this position?
This is a highly hands-on role that requires physical datacenter work. Candidates must be able to lift and move fifty or more pounds while performing tasks such as racking, stacking, and cabling.
What security clearance is required?
Candidates must be eligible to obtain and maintain an active U.S. Top Secret clearance. This requirement is necessary to support defense technology programs.
What kind of benefits are included?
Anduril offers top-tier benefits for full-time employees, including comprehensive and competitive coverage for health and recovery. Additionally, highly competitive equity grants are included in the majority of full-time offers.
The application is on the official Anduril Industries careers page.
Apply on company site