Lead DevOps Engineer

Lead DevOps Engineer taking ownership of Noxtua's infrastructure — a managed Kubernetes environment across multiple stages on sovereign EU cloud infrastructure (OTC), alongside self-hosted GPU servers. You'll set technical direction for a small DevOps team of 4–5 engineers while staying hands-on with the systems themselves.

What you'll do

• Own and optimize Noxtua's infrastructure across OTC and self-hosted GPU servers — ensuring efficient architecture, reliable operation, and cost control.
• Lead and grow a team of 4–5 DevOps engineers, setting technical direction, supporting their development, and having a strong ownership mindset.
• Operate the self-managed GPU server fleet — provisioning, driver installation, hardening, and connectivity via Ansible — and manage provider SLAs to keep heavy AI workloads running reliably.
• Build and maintain infrastructure automation using Infrastructure as Code (Terraform & Ansible).
• Run the container platform on Kubernetes, support teams with Docker, and keep services stable, accessible, and secure.
• Set up and maintain monitoring and alerting (e.g., Prometheus, Grafana) to ensure system reliability and performance.
• Develop and maintain CI/CD pipelines and collaborate with development and AI teams to automate deployments.

What you'll need

• Experience leading or mentoring a team, setting technical direction, and balancing hands-on operations with people responsibility.
• Managed a fleet of servers and understand the methodology behind it — including OS-level operations on real hardware and working with provider SLAs.
• Strong proficiency in Linux and Bash, plus a scripting language such as Python.
• Proven track record designing, operating, and cost-managing cloud-based architectures — ideally OTC, or transferable experience from AWS, Azure, or Google Cloud — with solid networking fundamentals.
• Strong focus on automating provisioning and configuration with Terraform and Ansible.
• Expertise in containerizing applications with Docker and running them at scale on Kubernetes.
• Able to set up and maintain monitoring/alerting tools (e.g., Prometheus, Grafana).

Nice to have

• Experience with GPU servers.

What they offer

• 100% remote work possible (given a German residence), other countries upon request.
• Flexible working hours.
• 26 days vacation + December 24th & 31st off, + 1 additional day per year of employment (up to 30 days).
• Equipment: Laptop (Lenovo or Mac), plus €1,000 net home office setup budget.
• Urban Sports Club Membership discounts depending on location.