DevOps Engineer (m/f/d) – basebox GmbH

Location: Utting am Ammersee (Bavaria), Germany — hybrid or remote (Germany-based) possible, with occasional travel to customer sites for GPU deployments Employment type: Permanent, full-time Start: Immediately / by arrangement Salary: €65,000–75,000 gross, depending on experience.

About basebox

basebox develops a sovereign AI platform for regulated organizations in the DACH region — particularly hospitals and municipalities. We focus on on-premises deployment, ISO 27001 certification, and full GDPR compliance. Our customers entrust us with critical infrastructure, so we're looking for someone who will share that responsibility with us and is firmly anchored in the team within the German/DACH legal framework.

Your tasks

• Operate and further develop our Helm-based deployment — making installations, upgrades, and releases stable and repeatable
• Own Kubernetes, GPU, inference, ingress, and identity infrastructure across the entire platform
• Contribute to a parameterized MCP Helm architecture for upcoming releases
• Prepare and operate customer and partner environments, including GPU deployments on-site
• Build reproducible deployment, recovery, and benchmarking processes in place of one-time manual installations
• Take part in an incident response rotation with platform-wide operational responsibility
• Work closely with our experienced DevOps freelancer during onboarding, gradually taking over and sharing platform responsibility

What we're looking for

Must have:

• Independent ownership — you can own a deployment or incident end-to-end without needing step-by-step direction
• Team collaboration — a team player who communicates proactively and surfaces blockers early rather than working in isolation
• Documentation as habit — you write runbooks and share what you learn so knowledge lives in the team, not just in your head
• Solid production experience with Kubernetes, Helm, and containerized deployments
• Knowledge of ingress and identity management (e.g. OAuth/OIDC)
• High reliability during incidents
• Professional German and English

Nice to have:

• Experience with GPU infrastructure / inference workloads
• Experience with customer or partner rollouts, ideally in regulated environments (healthcare, public sector)
• Interest in or experience building a Kubernetes operator (Go/Rust) — to help us move from Helm-only toward operator-managed lifecycle (upgrades, health, reconciliation)

What we offer

• Responsibility for a central platform with real impact in healthcare and the public sector
• Permanent employment
• A close, small team with short decision-making paths
• A structured onboarding alongside an experienced freelance colleague, so you build platform ownership together rather than carrying it alone from day one