Staff AI Infrastructure Engineer — Orchestration & Inference

Hamilton Barnes Associates Limited; San Francisco (CA); On-site; USD 213,000 - 288,000; Full time

Benefits offered by this job

Early-stage equity

Direct access to leadership

Job summary

Hamilton Barnes Associates Limited is seeking a Staff-level engineer to architect and evolve the AI infrastructure software stack for large-scale GPU workloads. You’ll drive orchestration, scheduling, and agentic operations across bare metal, Slurm, Kubernetes, and InfiniBand in a live production environment.

You’ll contribute to inference platforms, model serving, and high-utilization microservices, while building IaC tooling to automate deployment across thousands of servers.

Qualifications

• Strong systems software background at the intersection of distributed systems and infrastructure automation.
• Experience building orchestration, scheduling, or cluster management software at scale (Kubernetes, Slurm, or equivalent).
• Solid understanding of GPU infrastructure (InfiniBand, NVLink, NCCL) and related software considerations.
• Production engineering mindset with shipped software on real infrastructure.
• Proficiency in Go, Python, or Rust; Linux systems programming.

Responsibilities

• Build and evolve the core AI infrastructure software stack, including orchestration, scheduling, cluster management, and the agentic operations layer for placement, healing, and recovery.
• Contribute to the inference platform, including token gateway, microVM pooling, and model serving infra for high utilization and low latency.
• Work across bare metal, Slurm, Kubernetes, and InfiniBand environments; write software that abstracts complexity for customers while remaining debuggable by engineers.
• Develop IaC tooling to automate deployment at scale and track changes across 10,000+ servers.
• Integrate with NVIDIA inference microservices, custom model pipelines, and customer workloads for simple consumption of GPU infra.
• Contribute to the SRE tooling layer: monitoring, optimization, GPU utilization balancing, and automated remediation.

Skills

Distributed systems

Orchestration software

Kubernetes

Slurm

GPU infrastructure

Go

Python

Rust

Linux systems programming

InfiniBand

NCCL

NVLink

Go/Python/Rust

Job description

Hamilton Barnes Associates Limited is seeking a Staff-level engineer to architect and evolve the AI infrastructure software stack for large-scale GPU workloads. You’ll drive orchestration, scheduling, and agentic operations across bare metal, Slurm, Kubernetes, and InfiniBand in a live production environment.

You’ll contribute to inference platforms, model serving, and high-utilization microservices, while building IaC tooling to automate deployment across thousands of servers.