About Verda
At Verda, we're building a full-stack AI cloud, covering everything from data centers and hardware to our own cloud platform that the world's leading AI teams use to do serious AI work. We strive to make a positive mark on the world through the infrastructure we build and give leading teams a service they can truly depend on. Headquartered in Helsinki, we operate globally with offices in London and San Francisco. Join Verda while it's still being built - not once it's finished.
Why Verda
•Cash and equity compensation along with local benefits.
•40+ nationalities, with 6 different ones on the management team.
•A real chance to make an impact and work alongside world class engineers, researchers, and partners across the global AI ecosystem.
About the role
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.
Responsibilities
•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
Your key competencies
The ideal candidate has a solid Linux and systems foundation, writes real code, and is curious to learn the rest of the stack. You don't need to tick every box to apply.
•3+ years in SRE, infrastructure, platform or systems engineering, or equivalent hands-on experience from internships
•Solid Linux fundamentals and hands-on troubleshooting experience
•Hands-on Kubernetes experience, ideally in production
•Some experience with infrastructure automation or GitOps (for example Ansible, ArgoCD, Terraform, CI/CD) and observability tools (for example Prometheus, Grafana, VictoriaMetrics, OpenTelemetry)
•Experience writing tooling and automation in Go or Python
•Solid networking fundamentals: TCP/IP, DNS and load balancing
•You debug methodically in Linux, networking or distributed systems, and you're curious to learn the rest of the stack
•Ownership and a data-led approach: you follow an incident through to the lasting fix, and you can show the effect of your work
•Good judgement about when to automate, when to fix by hand and when to escalate
•Clear communication and a quick learner: you share progress, ask for help early, act on feedback and work well across teams
Nice to have
•GPU infrastructure: NVIDIA drivers, DCGM, InfiniBand or RoCE, and common GPU failure modes
•Bare-metal provisioning and hardware lifecycle: BMC/IPMI, Redfish and firmware management
•Linux performance analysis, or driver and kernel-level troubleshooting
•Tools from our stack: Rancher, kubeadm, Cilium, containerd, Harbor, Kyverno
•Building AI or agent-based tools for operations
•Experience at a cloud provider or running a large data-centre fleet
•Previous on-call or incident response experience for production systems
Practicalities
•Location: Helsinki, Finland
•Work mode: Hybrid, 3 days a week at the office
•Employment type: Full time and permanent
We're building fast and this role needs the right person behind it. There's no artificial deadline, but when we find who we're looking for, we move. Please submit your application through our Careers page. We don't accept applications sent by email.
