About the job

Are you a DevOps Engineer with a passion for infrastructure platforms? Do you want to work in a high-tech environment where research, innovation and education come together? Then we are looking for you!

TU Delft operates its own data centres and its own campus network. Deliberately. As a scientific institution, we believe that maintaining control over our digital infrastructure is not a matter of cost, but an essential prerequisite. You make sure that this infrastructure is not managed manually, but controlled through code.

Job Description

You will work on one of the largest and most distinctive IT environments in the Netherlands. More than 40,000 researchers, students and staff rely on it every day. Behind this environment is not a standard catalogue of services, but a campus network that simultaneously supports lecture halls, laboratories, quantum research and high-performance computing — all from data centres that we own and operate ourselves.

We are transforming this environment from manual operations to code. Not as a project with an end date, but as a way of working. Infrastructure as Code (IaC) is the standard, not the ambition. Everything goes through Git, everything is peer-reviewed, and what we build moves into production quickly so that we can learn from real feedback rather than assumptions.

Your work is at the heart of this transformation. You automate the deployment, configuration and registration of our data centre, network and platform services using tools such as Ansible, Terraform, NetBox, GitLab CI and Python. You help build the internal platform that enables other teams within the ICT Directorate to securely consume standardised infrastructure services themselves, rather than submitting a ticket and waiting.

What You Will Work On

Depending on your interests and strengths, your work will focus on areas including:

I. Establishing a reliable source of truth

NetBox is becoming our technical Configuration Management Database (CMDB). Network and infrastructure configurations should originate from it, rather than the other way around. That may sound less exciting than some of the other challenges, but it is the foundation for everything we automate next.

II. Connecting to our shared automation landscape

We are building an event-driven orchestration layer in which a request from a self-service portal automatically results in a deployed, registered and monitored service. From a portal request, to the desired state, to working infrastructure — with the resulting status fed back to the requester.

III. Compartmentalised research environments

For sensitive research, we build isolated environments driven by knowledge security requirements and the forthcoming Dutch Knowledge Security Screening Act (Wet screening kennisveiligheid). Segmentation, access control and demonstrable compliance are not secondary considerations here; they are part of the product itself.

IV. Rethinking the virtualisation layer

We are exploring Proxmox as an alternative to our current VMware stack. What this means for networking, storage and physical infrastructure remains an open technical question — one that you will help us explore, design and build.

V. Moving from monitoring to measurable reliability

Knowing that something is broken is not the same as knowing how reliable it is. Observability is one of the major areas we still need to develop further.

You will help shape our technical direction. You will have the freedom to make choices, initiate improvements and introduce new tooling — and we expect you to be able to explain and defend those choices to colleagues and stakeholders who may see things differently.