Go Kubernetes Engineer — On-Demand Metal Cloud (GPU)

Mistral.ai; Paris; On-site; EUR 100,000 - 160,000; Full time

Job summary

Mistral.ai is seeking an experienced backend infrastructure engineer to design, implement, and operate the Kubernetes-based platform that powers our sovereign AI cloud. You will own critical control-plane components, automate provisioning, and ensure reliability across on-prem and edge deployments.

You will work closely with product, design, and platform teams to ship scalable, production-grade systems and drive performance, security, and operability across the fleet of GPUs and compute

Qualifications

• Shipped production systems in Go.
• Run Kubernetes in anger; ideally built operators, schedulers or controllers.
• Experience running GPU workloads on Kubernetes or ready to own that domain quickly.
• Think like a developer who builds infrastructure, not just an operator.
• Extensive experience building backend or infrastructure systems.
• Collaborated with product, design and business teams to ship software used by real users.
• Demonstrates a low-ego, team-first mindset.

Responsibilities

• Design and ship the control plane architecture and implementations for the Kubernetes product.
• Own the reliability of the Kubernetes control plane and its KPI.
• Automate cluster provisioning to minimize manual tickets.
• Collaborate with Product Managers to set priorities and scope initiatives.
• Raise the engineering bar through code reviews and knowledge sharing across the Cloud Platform org.

Skills

Go

Kubernetes

Backend engineering

Infrastructure design

Linux

GPU workloads

Team collaboration

Tools

Kubernetes

Linux

Containers

Job description

Mistral.ai is seeking an experienced backend infrastructure engineer to design, implement, and operate the Kubernetes-based platform that powers our sovereign AI cloud. You will own critical control-plane components, automate provisioning, and ensure reliability across on-prem and edge deployments.

You will work closely with product, design, and platform teams to ship scalable, production-grade systems and drive performance, security, and operability across the fleet of GPUs and compute