Key Skills

Job Description

About Nostrion Nostrion builds independent European AI for organisations that work with sensitive information and operate in regulated sectors. We help professionals put their organisation’s knowledge to work while retaining control over their data, processes, and expertise. European ownership, infrastructure, and governance form the foundation of our approach. We combine open models with our own platform and work alongside domain experts to develop applications that fit their profession’s daily needs. Fidacta, our solution for the notarial sector, makes this tangible. It helps notaries and their teams search documents, obtain answers grounded in their own information, generate draft documents, and streamline workflows. These applications depend on a platform that handles sensitive customer data and AI workloads securely and reliably. Your Role As a Platform Engineer, you build and operate the technical environment behind our AI applications and customer environments. You bring together NVIDIA GPU capacity, Kubernetes, and AI model infrastructure to create a reliable, secure, and manageable platform. You deploy models, provision customer environments, and automate releases. Your work spans the full stack of infrastructure: GPU resources, networking, storage, model serving, monitoring, and access management. You take responsibility for how models run in production: their performance under load, their observability, and how they can be updated safely. You investigate bottlenecks, balance performance against capacity and cost, and help engineers bring AI applications into production with confidence. What You’ll Be Doing Build, operate, and improve Kubernetes environments for AI workloads and customer applications. Make NVIDIA GPUs available to workloads and provide visibility into GPU utilisation and capacity. Automate AI and LLM deployments, including configuration, versioning, and controlled updates. Configure and optimise model-serving environments for latency, throughput, and efficient GPU memory usage. Provision customer environments with appropriate isolation of workloads, data, network traffic, and access. Develop Infrastructure as Code and reusable configurations for consistent, reproducible environments. Build and improve CI/CD and GitOps processes for platform components and model deployments. Implement monitoring, logging, and alerting for infrastructure, GPUs, model endpoints, and customer environments. Improve availability, backups, recovery procedures, and capacity planning. Investigate incidents, resolve root causes, and implement lasting improvements. Work with AI and software engineers to turn experiments into stable production services. Document technical decisions, configurations, and operational procedures clearly. What You Bring You are a practical engineer who enjoys understanding how systems work and takes responsibility for their reliability. You can build and operate infrastructure while keeping the needs of developers and users in mind. Experience as a Platform Engineer, Infrastructure Engineer, DevOps Engineer, SRE, or in a similar role. Strong knowledge of production Kubernetes, including deployments, scheduling, networking, storage, and access control. Hands-on experience with NVIDIA GPUs in Linux and container environments. Experience running or supporting AI and LLM workloads, using tools such as vLLM, NVIDIA Triton, TensorRT-LLM, or similar technologies. Experience with Infrastructure as Code, such as Terraform or OpenTofu, and deployment tooling such as Helm. Knowledge of CI/CD, ideally including GitOps with tools such as Argo CD or Flux. Experience monitoring and troubleshooting distributed systems. Scripting or programming skills in languages such as Python, Bash, or Go. An understanding of network security, secrets management, and isolation between customer environments. A careful, independent approach and the ability to explain technical trade-offs clearly. Bonus Points If You Have Experience with NVIDIA GPU Operator, NVIDIA Container Toolkit, and GPU monitoring with DCGM. Knowledge of GPU sharing, MIG, multi-GPU inference, and GPU workload scheduling. Experience optimising LLM inference through batching, quantisation, or KV-cache management. Experience with private cloud, on-premises infrastructure, or environments with restricted external connectivity. Familiarity with model registries, artifact storage, and reproducible model releases. Experience improving networking and storage performance for large models and datasets. Experience in secure or regulated environments where changes and access must be demonstrably controlled. Your Team You work closely with AI Engineers, Software Engineers, Product, and colleagues supporting customer implementations. Together, you ensure that new models and applications become available safely and reliably. You help engineers work independently through clear deployment processes, useful monitoring, and dependable platform services. You use lessons from production to improve both the infrastructure and the developer experience. What We Offer Ownership of the infrastructure behind our AI applications and customer environments. A hands-on technical role working with NVIDIA GPUs, Kubernetes, and modern AI and LLM technologies. The opportunity to shape architecture, automation, and platform standards. Direct collaboration with the engineers developing models and applications. A competitive salary and attractive employment conditions. The opportunity to help build sovereign European AI for professional use. How to Apply Send us your CV and briefly describe a platform or technical environment you helped build: What ran on it, and what were you responsible for? What challenge did you solve involving Kubernetes, GPUs, model deployments, or reliability? What technical trade-off did you make, and what did you learn? Apply through Frank Poll. Learn more at nostrion.com and fidacta.ai.

Core Responsibilities

Build and operate secure, reliable Kubernetes infrastructure for AI workloads and customer environments, including GPU capacity, model serving, networking, storage, monitoring, and access management. Automate deployments and releases, improve reliability and performance, investigate incidents, and collaborate with AI and software engineers to bring applications into production.

Requirements

Candidates should have experience in a platform, infrastructure, DevOps, or SRE role, strong production Kubernetes knowledge, and hands-on experience with NVIDIA GPUs and AI or LLM workloads. They should also be familiar with Infrastructure as Code, deployment automation, monitoring distributed systems, scripting or programming, and security and isolation practices.

Benefits

• Competitive Salary
• Attractive Employment Conditions