Erhalte mehr Antworten von Arbeitgebern
Versende in nur wenigen Minuten einen passgenauen Lebenslauf.
Fuchs & Eule in Berlin seeks an experienced infrastructure engineer to own and evolve our cloud and on‑prem hybrid stack. You will manage AWS (EKS, RDS, EC2, IAM) and Hetzner, drive IaC with Terraform/OpenTofu, and keep Kubernetes healthy across environments.
You will enable data scientists and engineers by building CI/CD pipelines, secure by design practices, monitoring, and incident response support. This role prizes pragmatic, collaborative problem solving and sharing of reusable templates
Own and Evolve Our Infrastructure – You take ownership of our AWS and Hetzner environments and evolve them further – everything as Infrastructure as Code, running, scaling, and upgrading our Kubernetes clusters along the way
Operate Our Platforms – You keep our webapps, self-hosted tools, and ML stack running in stable, secure production environments, with solid monitoring, logging, and alerting across systems – including observability for LLM-based workloads
Keep Us Secure and Compliant– You handle patching, secrets management, access control, and backups, and support audits and certification efforts as part of your routine work
Build Paved Roads for Engineers and Data Scientists – You provide the containerization, CI/CD, and deployment tooling that let both teams ship and run their own services – from webapps to RAG pipelines – enabling rather than taking over, as deployment stays a shared responsibility
Production-Grade Cloud Experience – You bring strong hands‑on AWS experience (EKS, RDS, EC2, IAM, networking) and are open to hybrid setups including bare‑metal/VPS providers like Hetzner
IaC and Kubernetes as Your Craft – Infrastructure as Code is your default working mode (OpenTofu/Terraform), backed by solid scripting skills (e.g. Bash, Python) and confidence troubleshooting Kubernetes in production
Fluent in the ML/LLM Ecosystem – You're conceptually familiar with RAG architectures, vector databases (e.g. Qdrant), and ML tooling (e.g. MLflow, Langfuse) – enough to speak the data scientists' language and support their infrastructure needs; deep hands‑on ML engineering is explicitly not required
Enabler Mindset, Pragmatic Approach – You share knowledge through pairing, reviews, and reusable templates, explain trade‑offs clearly to any audience, prioritize autonomously, and choose "good enough" over over‑engineered