Senior DevOps / Platform Engineer

Senior DevOps / Platform Engineer Nimbus Data Systems — AI Venture Builder Remote · Distributed team (Europe + KSA) About Nimbus Data Systems Nimbus Data Systems is a Saudi-based AI venture builder. We design, build, and operate AI products end to end. We run a lean, distributed engineering team that ships fast and favours self-hosted, open-source tooling we control over opaque SaaS. We're building serious infrastructure for the Kingdom and beyond, and we need a platform engineer who treats reliability, security, and developer experience as first-class products. The Role You'll join the platform team that owns the infrastructure our products run on: a multi-cloud, multi-region Kubernetes estate spanning GCP (Dammam) and OCI (Jeddah). This is a hands-on senior role — you'll architect the GitOps and delivery pipelines, harden our service mesh and secrets posture, and make sure the developers building on top of your work barely have to think about infrastructure. You won't be a one-person army: you'll work alongside other platform and application engineers, sharing ownership and on-call rather than carrying it all alone. If you like owning systems rather than tickets, and you want a stack with no legacy baggage, this is for you. What You'll Work On Multi-cloud, multi-region: GCP and OCI across a multi-project layout, including a credible, tested failover story between the two. Infrastructure as Code: Terraform for cloud resources; Helm and Kustomize for Kubernetes workloads. Everything reproducible, reviewed, and versioned. Orchestration: Kubernetes across both clouds — Helm-packaged workloads, HPA-driven autoscaling, sensible resource and capacity management. Service mesh: Istio for traffic management, mTLS, and observability at the network layer. GitOps & progressive delivery: ArgoCD as the source of truth for cluster state, Argo Rollouts for canary and blue-green deploys, GitHub Actions for CI. Secrets & identity: HashiCorp Vault as the secrets backbone, External Secrets Operator to sync into clusters, and Keycloak for SSO and identity across our services. Data platform: Cloud SQL for PostgreSQL 17/18 and Memorystore for Redis 7, supporting application teams who use Drizzle ORM with a forward-only migration discipline. Observability: The full Grafana LGTM stack (Loki, Grafana, Tempo, Mimir) for logs, dashboards, traces, and metrics. Responsibilities Design and operate the production and DR Kubernetes platform, owning availability, scaling, and the GCP→OCI failover path. Build and maintain GitOps delivery pipelines that let teams ship safely with progressive rollout and fast rollback. Manage the secrets and identity layer (Vault, ESO, Keycloak) and keep our security posture tight across both clouds. Codify all infrastructure in Terraform/Helm/Kustomize and drive consistency across projects and environments. Run the observability stack and turn signals into actionable alerts and SLOs — not dashboard noise. Partner with application engineers on the data layer: connection management, performance, and a clean forward-only migration workflow. Improve developer experience: reduce toil, automate onboarding, and make the platform self-service where it makes sense. What You Bring Must have Strong production experience operating Kubernetes at scale, including Helm and autoscaling. Deep Terraform skills and a real IaC discipline. Hands-on GitOps experience with ArgoCD (or equivalent) and progressive delivery patterns. Solid grounding in secrets management and identity — Vault, ESO, Keycloak or comparable tools. Experience with managed PostgreSQL and Redis and the operational concerns around them. Practical experience with observability , ideally the Grafana ecosystem (Loki/Tempo/Mimir/Grafana). A security-first mindset and comfort owning incidents end to end. Nice to have Multi-cloud experience, especially GCP and OCI together. Hands-on Istio or another service mesh in production. Experience designing and testing DR / failover across regions or clouds. Familiarity with Argo Rollouts specifically. Background supporting AI/ML or agentic workloads in production. Comfort working in a distributed, async, fast-moving team. How We Work Remote-first and distributed across European and KSA time zones — we expect meaningful overlap with the team's core hours. Open-source and self-hosted by default; you'll have real ownership of the tools you run. High autonomy, low bureaucracy, and a stack you can shape rather than inherit. Apply If this sounds like your kind of problem, reach out with your CV or profile and a few lines on a platform you've built or rescued. We'd love to hear from you.