AI Infrastructure Lead

Outsourceit; San Francisco (CA); On-site; USD 120,000 - 170,000; Full time

Job summary

Outsourceit is seeking an AI Infrastructure Lead to design and operate innovative GPU infrastructure for enterprise AI workloads. This role requires a minimum commitment of 6 months and involves working closely with the CTO and a team of engineers.

The ideal candidate will have proven experience with large-scale GPU systems, distributed training, and containerised workloads, alongside expertise in AWS, GCP, or Azure services. Full-time availability and participation in an on-call rotation are required.

Qualifications

• Proven experience designing and operating large-scale GPU infrastructure and model serving systems.
• Deep knowledge of distributed training, inference optimisation, and containerised workloads.
• Hands-on expertise with AWS, GCP, or Azure AI/ML services and Kubernetes.

Responsibilities

• Own the design and operation of the GPU cluster management layer.
• Lead the model serving pipeline and low-latency routing system.
• Make architectural decisions that affect thousands of enterprise clients.

Skills

Large-scale GPU infrastructure design

Model serving systems

Distributed training

Inference optimisation

Containerised workloads

AWS services

GCP services

Azure AI/ML services

Kubernetes

Monitoring

Alerting

Incident response

Job description

Parallax Systems is building the next generation of inference infrastructure for enterprise AI workloads. We need an AI Infrastructure Lead to own the design and operation of our GPU cluster management layer, model serving pipeline, and low-latency routing system.

You will work directly with the CTO and a team of four senior engineers. This is a high-ownership role in a fast-moving environment — you will make architectural decisions that affect thousands of enterprise clients.

Core Requirements

• Proven experience designing and operating large-scale GPU infrastructure and model serving systems.
• Deep knowledge of distributed training, inference optimisation, and containerised workloads.
• Hands‑on expertise with AWS, GCP, or Azure AI/ML services and Kubernetes.
• Strong background in monitoring, alerting, and incident response for critical AI systems.

Rolling contract with 6‑month minimum commitment.

Work load 40 HRS/WK

Full‑time availability required. On‑call rotation included.