GPU ML Systems Engineer — AI Compute Orchestration (Remote Considered)

SkyPilot; San Mateo (CA); On-site; USD 180,000 - 280,000; Full time

Benefits offered by this job

Equity

Medical/Dental/Vision coverage

Autonomy & ownership

Job summary

SkyPilot is seeking an engineer to own the GPU and ML-systems layer that powers frontier AI workloads. You will shape accelerator scheduling, utilization, and the serving path across clouds and Kubernetes, delivering a fast, cost-efficient AI stack.

You will deepen integrations with vLLM, PyTorch, Slime and related frameworks while enabling scalable training, inference, and multi-cluster serving. You’ll work with a team moving cutting-edge AI compute forward at scale.

Qualifications

• Hands-on experience with GPU or accelerator systems and ML training/inference infrastructure.
• Strong Python, and comfort with systems-level and GPU-adjacent details.
• Experience operating large-scale training or high-throughput inference in production.

Responsibilities

• Own GPU scheduling, utilization and health across clouds and Kubernetes with real-time monitoring and automatic recovery.
• Build optimizations for training and serving: large-scale pre-training, checkpointing, and multi-cluster serving.
• Deepen integrations with vLLM, PyTorch, CUDA and the ML frameworks used for pre-training and high-throughput inference.

Skills

GPU systems

Python

ML training/inference

Systems-level programming

Kubernetes

Tools

vLLM

PyTorch

CUDA

Slime

Kueue

KServe

Job description

SkyPilot is seeking an engineer to own the GPU and ML-systems layer that powers frontier AI workloads. You will shape accelerator scheduling, utilization, and the serving path across clouds and Kubernetes, delivering a fast, cost-efficient AI stack.

You will deepen integrations with vLLM, PyTorch, Slime and related frameworks while enabling scalable training, inference, and multi-cluster serving. You’ll work with a team moving cutting-edge AI compute forward at scale.