Site Reliability Engineer

Tehtävä

As a Site Reliability Engineer you keep Verda's GPU cloud running and make it more reliable every quarter. This role is for engineers with a few years of production experience. You start with well-scoped work alongside senior SREs and grow into owning parts of the platform from build to production. You work hands-on across the systems that keep our data centres serving customers: deployments, Kubernetes platforms, secrets, observability, and the tooling that connects our bare-metal GPU servers to the services our customers use. At Verda, reliability is part of the product. You measure it in SLOs, incidents, pages and customer credits, and you write the code and automation that bring those numbers down. You work closely with our automation, networking, Linux, hardware and data-centre operations teams. Once you're ready, you join our paid primary/secondary on-call rotation, paired with experienced engineers while you learn. Run and improve the services that keep our data centres operating, from deployment pipelines to secret provisioning Contribute to SLIs and SLOs for our core platform services Take part in on-call: respond to incidents, lead smaller ones as you gain experience, contribute to blameless reviews and follow through on the fixes you own Diagnose failures methodically across the stack, from GPU and server hardware through Linux and networking to Kubernetes and the services on top Build automation that removes repetitive operational work, including AI-assisted and agent-based tooling where it helps Make deployments safer with release criteria, readiness checks, progressive rollouts and fast rollback Build monitoring into the work you ship and extend our observability (metrics, logs, traces and alerts), so we find problems before customers report them Contribute to capacity planning, failover and redundancy design for a fast-growing GPU fleet Raise reliability risks early with clear options, and log them in the infrastructure risk register Write runbooks and documentation, and help your peers through code reviews