$262,700 - $355,400 USD yearly
Originally posted October 21, 2026 by the employer.
About the role
As a Principal Engineer on the AI Compute Infra team, you will design, build, and operate large-scale infrastructure for AI training, fine-tuning, evaluation, and inference. You will guide work across Kubernetes clusters, accelerator enablement, workload scheduling, high-performance networking, storage, and capacity management.
What you'll do
- Build and operate Kubernetes clusters while improving workload scheduling, topology-aware placement, capacity use, and recovery.
- Enable new CPU and GPU systems by integrating and validating drivers, networking, storage, monitoring, and health checks.
- Investigate performance and reliability issues across applications, cloud infrastructure, clusters, and hardware, turning findings into lasting improvements.
- Partner with AI teams to understand their workloads and automate cluster provisioning, upgrades, monitoring, and maintenance around their needs.
What you'll need
- 12+ years of experience building or operating cloud, compute, HPC, or distributed infrastructure in a production environment.
- Programming experience in Go, Python, or another systems language.
- Practical knowledge of Kubernetes, containers, Linux, networking, and storage.
- Experience supporting GPU, accelerator, or distributed machine-learning workloads.
- Demonstrated ability to troubleshoot complex systems and communicate across technical disciplines.
Nice to have
- Familiarity with Kubernetes scheduling, operators, quotas, or resource management.
- Experience with NVIDIA technologies including CUDA, NVLink, NVSwitch, NCCL, EFA, or DCGM.
- Knowledge of AWS EKS, Terraform, Argo CD, Helm, Prometheus, or Grafana.
- Familiarity with frameworks such as PyTorch, Ray, vLLM, SGLang, or TensorRT-LLM, or experience qualifying accelerators and tuning distributed workloads.
Skills: Kubernetes, GPU, CUDA, Linux, cloud infrastructure
More semiconductor jobs: Arm Holdings jobs
Tracked in our index since 29 Oct 2026. This is our observation date, not the employer posting date. Specialty: Infrastructure/Platform Software (Infrastructure Platform). 1067 active roles in this specialty at 97 employers in our index, as of 20 Nov 2026.
How we counted: postings read from employer career sites. How we count