DevOps Engineer - KubeOps Team

Advantage

- Cluster autoscaling in production
- Cost / capacity optimization at fleet scale — Karpenter consolidation, bin-packing, spot strategy
- Linux internals and node-level debugging — when a Kubernetes problem turns into a Linux problem, you can take it from there
- Track record of leading projects end-to-end: scoping, execution and delivery.

At Wix, we believe our best work happens together. Our work model is fully in person, with 5 days a week from our office. Flexibility remains a core value at Wix and special requests are handled thoughtfully at the team level.

Responsibilities

Master the Cluster: Manage the lifecycle of our large-scale Kubernetes environments, without disrupting the thousands of services running on it.; Build the Platform: Own and evolve the cluster building blocks — Kubernetes Architecture, Karpenter configurations, addons, networking, autoscaling. Where it makes sense, platformize them so other infra teams can consume them cleanly; Pioneer AI-Driven Ops: Lead the charge in integrating AI into our ecosystem—building LLM-powered tools that accelerate investigation, planning, and coding for the whole team.; Full-Stack Ownership: You’ll take ambiguous tasks and transform them into high-quality outcomes, owning every decision along the way.

Requirements

+2 years in managing infrastructure, operating production systems at large scale; 2+ years running production Kubernetes — EKS preferred with a strong understanding of core control-plane and cluster components, with hands-on experience debugging, upgrading, and tuning autoscaling and cluster networking; 2+ years with a major cloud provider (AWS preferred) — solid grasp of core compute, networking, and IAM.; 2+ years with Infrastructure-as-Code, writing and maintaining reusable modules (Terraform, Pulumi, Crossplane, etc.); Familiarity with observability at scale — Prometheus, Grafana, alert design; Experience building internal tooling in Go or Python; Hands-on experience with AI-assisted development tools and agents (Claude Code, Codex, Cursor) for knowledge gathering, debugging, coding, planning, and design — with the ability to integrate these into concrete workflows that demonstrably accelerate your work.