AI / ML Infrastructure Engineer
Summary
Designs, scales, and optimizes the GPU infrastructure behind large-scale generative AI/ML models in production — building NVIDIA GPU clusters for training and low-latency inference with Triton, vLLM, and TensorRT. Core stack: PyTorch, CUDA, Kubernetes/Docker, AWS/GCP, and Terraform, in a hybrid/on-site role in New York or San Francisco.
Location: New York, NY or San Francisco, CA (Hybrid / On-site)
Employment Type: W-2 Contract or Full-Time (Must be authorized to work on W-2 in the US without sponsorship)
Experience: 3+ Years
Job Summary
We are seeking an AI / ML Infrastructure Engineer to join our client's high-performance engineering team. You will design, scale, and optimize the underlying infrastructure powering large-scale generative AI and machine learning models in production.
Key Responsibilities
Architect and scale high-performance GPU clusters (NVIDIA hardware) for model training and low-latency inference.
Optimize model serving pipelines using Triton Inference Server, vLLM, and TensorRT.
Collaborate with data scientists and MLOps engineers to streamline distributed training workflows.
Monitor, troubleshoot, and optimize cloud infrastructure costs and performance bottlenecks.
Requirements
Minimum 3 years of hands-on experience in ML infrastructure, systems engineering, or DevOps roles.
Strong proficiency in PyTorch, CUDA ecosystem, and container orchestration (Kubernetes, Docker).
Deep experience with cloud providers (AWS, GCP) and Infrastructure as Code (Terraform).
Valid W-2 work authorization in the United States (no C2C or cross-border arrangements).
Sound like a fit?
Apply now and our recruiting team will get back to you.