AI Infrastructure Engineer - GPU as a Service

Salary: $180,000 - 280,000 per year Requirements:

• We are looking for at least 4 years of experience shipping production systems, including hands-on work with GPU or ML infrastructure.
• We need strong Python skills plus proficiency in at least one of Go, C++, or Rust.
• We expect practical experience with CUDA, NCCL, or similar NVIDIA GPU technologies in production settings.
• We prefer experience with Kubernetes, Linux, and high-performance networking.
• We value candidates who can thrive in a small team and take ownership of problems from start to finish.
• Bonus experience includes neocloud, GPU cloud, or HPC scheduler environments.
• Additional advantage if you have worked with LLM serving tools such as vLLM, TensorRT-LLM, or Triton.
• Familiarity with multi-tenant isolation, billing metering, or quota systems is a plus.

• We build and operate the control plane for GPU allocation, scheduling, and multi-tenant service delivery.
• We deploy inference and training workloads on our clusters using CUDA, containers, Kubernetes, or comparable tools.
• We own the reliability, utilization, and performance of our GPUaaS platform.
• We collaborate with site operations and customers on cluster bring-up, networking, and capacity planning.
• We develop the APIs and internal tooling that our customers and teams rely on.

• Cloud
• CUDA
• Kubernetes
• LLM
• Linux
• Python
• Rust
• vLLM
• DevOps

More:

We are DigiPower X (NASDAQ: DGXX), a Bay Area-based company offering a full-time role in an on-site or hybrid environment. This production-focused position builds the software behind our GPU-as-a-Service platform, bringing our NVIDIA B300, Vera Rubin, and related GPU systems to customers as a service. The compensation range is $180,000 to $280,000 base salary, depending on scope and experience, plus equity.

last updated 33 week of 2026