Autonomous GPU Infra & Fleet Automation Engineer

Together AI; Amsterdam; On-site; EUR 90,000 - 150,000; Full time

Benefits offered by this job

Competitive health insurance

Mental health support

Dental and vision insurance

Income protection & retirement

Pre-tax benefits

Job summary

Together AI in Amsterdam is building and operating one of the world’s largest GPU fleets for frontier model training and inference. You’ll design automation to provision, validate, deploy, upgrade, repair, and retire GPU clusters with minimal human intervention.

You will help create AI infrastructure agents, fleet intelligence platforms, and internal tools to manage tens of thousands of GPUs autonomously, collaborating across hardware, networking, and AI teams.

Qualifications

• 3+ years building distributed systems or large-scale backend software
• Strong systems thinking across hardware and software
• Experience with Linux, Kubernetes, Terraform, Ansible
• Proficient in Python, Go or Rust
• Experience building automation platforms or developer infra
• Interest in GPU infrastructure and AI training/inference clusters

Responsibilities

• Design and build fleet automation systems for GPU clusters with minimal human intervention
• Develop AI Infrastructure Agents for deployment, failures, incident triage, and remediation
• Create Fleet Intelligence platforms monitoring hardware health and performance
• Build software that maximizes GPU availability, utilization, performance, and reliability
• Collaborate with hardware, networking, platform, and AI teams to push AI infra limits

Skills

Distributed systems

Automation

Python

Go

Rust

Linux

Performance

Scalability

Tools

Kubernetes

Terraform

Ansible

InfiniBand/RoCE

NVLink/NVSwitch

Job description

Together AI in Amsterdam is building and operating one of the world’s largest GPU fleets for frontier model training and inference. You’ll design automation to provision, validate, deploy, upgrade, repair, and retire GPU clusters with minimal human intervention.

You will help create AI infrastructure agents, fleet intelligence platforms, and internal tools to manage tens of thousands of GPUs autonomously, collaborating across hardware, networking, and AI teams.