AI platform engineer

Thales Group; Amsterdam; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Hybrid work environment

Competitive salary and 40 vacation/yr

Job summary

Thales Group in the Netherlands seeks a DevOps/Platform Engineer to advance our GenAI on-prem ecosystem. You will work in a highly secured environment, building, hosting and managing AI infrastructure entirely in-house on GPU clusters, with air-gapped data flows and strict security rails.

You will deploy AI tooling, manage RAG pipelines, and enable local model routing and multi-agent tooling within the closed network. Fluency in Dutch and English is required, and security screening is mandatory.

Qualifications

• On-Premise and DevOps experience in Linux environments and on-premise data centers.
• Strong programming skills in Python and experience building and integrating local APIs.
• GenAI knowledge: vector indexing, embeddings, and large language models; open-source tools experience preferred.
• Fluent Dutch and English, spoken and written.
• Security screening: national security screening required.

Responsibilities

• AI Container Application Management: deploy, update and maintain AI tooling in on-premise container platforms.
• Air-Gapped RAG and Codebase Infrastructure: build and maintain on-premise retrieval infrastructure.
• On-Premise Model Routing and Lifecycle: configure tooling for local model serving and routing across tasks.
• Agent Orchestration and Tooling: develop local multi-agent frameworks and memory interfaces.
• Air-Gapped Observability: implement local monitoring and tracing dashboards within the network.
• Strict AI Safety and Security Rails: implement guardrails to prevent prompt injections and ensure data compartmentalization.

Skills

Linux

Python

APIs

Security

Tools

Kubernetes

Docker

Job description

Описание

Are you a DevOps or Platform Engineer ready to make the leap into Generative AI, within one of the most highly secured software environments in the Netherlands? At Thales Netherlands, we are building our own advanced AI ecosystems in Hengelo, running in fully air-gapped infrastructures. As an AI Platform Engineer, you will ensure our software engineers and systems can safely use local language models, advanced multi-agent networks and systems for context-based information retrieval. Since we cannot use public cloud APIs, we build, host and manage everything ourselves on-premise on our high-performance GPU clusters.

Задачи

AI Container Application Management: You are responsible for deploying, updating and maintaining critical AI utilities and developer tools within the on-premise container orchestration platforms. You ensure optimal performance, resource allocation and high availability for the development teams.

Air-Gapped RAG and Codebase Infrastructure: You build and maintain the on-premise Retrieval-Augmented Generation (RAG) infrastructure. You ensure applications can securely and quickly retrieve context from our internal codebases via local data pipelines, text segmentation (chunking) and vector indexing.

On-Premise Model Routing and Lifecycle: You configure the tooling for local model serving and help build a flexible routing layer. This allows applications within the closed network to dynamically switch between different local models based on the complexity of the specific task.

Agent Orchestration and Tooling: You support the creation of local frameworks for multi-agent systems (such as specialized coding, architecture and debugging agents). You help integrate local memory systems and interfaces that allow agents to call external tools.

Air-Gapped Observability: You implement monitoring and tracing dashboards that run entirely locally within the network. These let you measure token consumption, processing time (latency) and infrastructure costs per team or user, without any data leaking outside.

Strict AI Safety and Security Rails: Together with our security experts, you implement hard guardrails and security infrastructures to prevent malicious input (prompt injections) and ensure agents strictly adhere to data compartmentalization and authorization levels.

Требования

On-Premise and DevOps Experience: Good experience with Linux environments, containerization and container orchestration in an on-premise data center setting.

Software Engineering: Strong programming skills in Python and experience building and integrating local APIs.

GenAI Interest and Conceptual Knowledge: You understand the theory behind vector indexing, embeddings and how large language models (LLMs) work. Preferably, you have already experimented with these on a hobby or project basis using open-source tools.

Languages: Excellent command of both Dutch and English, spoken and written.

Security Screening: Since you will work on defense-related projects, successfully completing a national security screening (Certificate of No Objection) is required.

Будет плюсом: Experience with other systems languages or scripting is a plus

Условия

A unique opportunity to pioneer state-of-the-art GenAI technology in a high-security enterprise environment. A flexible hybrid working environment with a healthy balance between working from home and collaborating at the High Tech Systems Park in Hengelo. Excellent employment conditions under the Metalektro Collective Labour Agreement, including a very competitive salary and 40 vacation days per year (27 vacation days + 13 ADV days). Room and budget to further develop yourself into an expert in AI infrastructure.