Site Reliability Engineer (SRE)

About the role

We own our hardware. As our SRE, you keep the GPUs serving, the deploys safe, and the on-call calm. You'll work closest to the metal — our inference runs on owned servers across six regions (Germany, Switzerland, California, Dubai, London and Singapore), with our European hub in Frankfurt.

What you'll do

• Run and harden our self-hosted infrastructure — containers, networking, GPU nodes.
• Build CI/CD and zero-downtime deploys for the model and the apps.
• Own monitoring, alerting and incident response; make on-call rare.
• Tune the boundary between cost and reliability for owned hardware.

Why it matters

Our promises about privacy and a free tier rest on infrastructure we control. You keep that foundation solid.

What we're looking for

• 4+ years in SRE / DevOps / platform roles.
• Strong Linux, Docker, and networking fundamentals.
• Comfortable with infrastructure-as-code and CI/CD pipelines.
• Calm under incidents; you write the runbook after.

Nice to have

• GPU / CUDA host operations.
• Experience running bare-metal rather than only cloud.