Intric builds a sovereign AI platform for organisations that need to keep their data close: public-sector bodies, municipalities and critical infrastructure. It’s a small team, which means the work is real, the decisions matter and you’ll get close to the systems that actually keep the product running.
As the second person on the platform team, you’ll work directly alongside the lead on day-to-day operations, investigations and larger infrastructure projects. This is an early-career role. You won’t be expected to arrive knowing the whole stack, but you will be expected to have solid foundations, learn quickly and take on more ownership as your understanding grows.
The platform is largely self-hosted and open source. Kubernetes, k3s, Helm and Argo CD sit at the centre of it, alongside Docker and Dokploy environments that still need to be operated and migrated. Around that are Vault, Tailscale, Terraform, GitHub Actions, PostgreSQL, Redis, RabbitMQ, S3-compatible storage, Weaviate and a full OpenTelemetry and Grafana observability stack. Parts are sparsely documented and some of the problems won’t have a ready-made runbook. The learning curve is steep.
What you’ll do
Get hands-on with the whole platform. Learn how our managed, dedicated and on-premises environments fit together. Make changes to Kubernetes manifests, Helm charts, Argo CD applications, Terraform and CI/CD workflows. Early changes will be scoped and reviewed closely; over time, you’ll be expected to drive routine work yourself.
Take ownership of observability over time. Build a clear picture of what we can see across metrics, logs, traces, dashboards and alerts. Improve what’s missing, reduce noise and make incidents easier to understand. The stack includes OpenTelemetry, Grafana, Mimir, Loki and Tempo.
Take on the day-to-day operational load. Triage alerts, investigate failing workloads, make routine configuration changes and verify that deployments and backups are doing what we expect. The goal is for you to handle documented and familiar issues without needing to escalate, while escalating unfamiliar or high-risk problems early.
Turn repeated work into automation. If something has to be done twice, look for a way to make the third time safer and faster. That might mean a shell or Python script, a GitHub Actions workflow, a better alert or a proper runbook.
Work across the application boundary. The workload is a Python and FastAPI backend with workers, RabbitMQ consumers and several stateful dependencies. You won’t be hired as a backend engineer, but you will learn enough about the application to follow a problem from ingress to API, queue, consumer and data store.
Document what you learn. When the documentation runs out, go looking for the answer. Once you find it, leave the system clearer for the next person.
Who you are
You’re comfortable in a Linux environment and want to understand what the machine is actually doing. You understand the basics of processes, permissions, networking, DNS, HTTP and TLS. You don’t need to know everything from memory, but you need enough of a model to investigate when something breaks.
You’ve used Git and Docker, and you can automate a small task in shell, Python or another scripting language.
You approach problems methodically. You gather evidence, form a hypothesis, test it and explain what you found. You go looking for answers when the documentation runs out, but you don’t hide blockers or guess when the blast radius is large.
You can move between investigations, operational work and implementation without losing track of what matters. You can fill in reasonable gaps in a brief yourself, while checking assumptions that could affect production.
An infrastructure environment that isn’t built entirely around AWS or GCP managed services strikes you as more interesting, not more limiting.
You might be a recent graduate or have built equivalent experience through projects, a homelab or early professional work. We care more about the foundations you have and how you learn than whether you already know our particular stack.
Excellent communication skills are needed for this role because collaboration with the Forward Deployed Engineering and Product teams is part of the day to day work.
What you don’t need
You don’t need professional experience with every technology listed above.
You don’t need to be a Kubernetes expert on your first day.
You don’t need to have operated a production platform before.
You do need to be curious, careful and ready to become independently useful faster than you would in a heavily layered engineering organisation.
About Intric
Europe's public sector is facing its biggest challenge in generations. A shrinking workforce must deliver more services to more people, and the gap grows every year. Intric exists to close it. We bring modern AI to the organisations that keep society running, and in doing so, we're defining how public services will work for decades to come.
We value ownership, high standards, and thoughtful execution. Titles matter less than contribution. If you see something that can be improved, you take initiative. If something is unclear, you ask questions. If a customer faces a challenge, you lean in.
We combine ambition with collaboration. We move fast, but we care deeply about doing things right. As we continue scaling across Europe, every team member plays a meaningful role in shaping both the company and the culture.
