Usamos cookies técnicas y el registro de visitas anonimizado de Zorky. Yandex Metrica (los datos se guardan en servidores en Rusia) solo se activa con tu consentimiento. Nunca vendemos datos. Más sobre cookies y estadísticas.
Zorky está en modo de prueba: todo es gratis. Los empleos y las páginas de empresas están abiertos a todos; los enlaces a las ofertas y tu selección personal se abren tras registrarte. Regístrate →
← Todos los empleos

Senior Software Engineer | Kubernetes Automation

Senioroffice6500 EUR · periodo no indicadoEstonia; Greece; Latvia; Montenegro; Slovakia, SKhace 3 sem.
Análisis del mercado
📊 DevOps / SRE: salarios y demanda en el mercado
Stack
argocdawsazurebacklogci/cdclickhousecloudgcpgithubgitlabgitopsgo
Detalles
Inglés
B2, obligatorio
Prácticas
CI/CD, GitOps
Para postular
verificación de antecedentes
Mostrar empresa y enlace
Gratis, tras iniciar sesión.
Postular
Abre la empresa y el enlace de arriba: tu candidatura va directamente al sistema de selección de la empresa, sin intermediarios.

Descripción

Sobre el puesto

What you'll be solving A pod is mid-request. The node it's running on is about to disappear - Spot is pulling the instance, or Autoscaler has flagged the node as underutilized and marked it for removal. The safe move is to wait.

The profitable move is to act now. Every team in this pipeline is built to close that gap: make the aggressive call and still guarantee nothing breaks. That guarantee is the hard part.

Deciding which nodes to kill, which pods to move, and which storage volume to resize is a genuinely hard optimization problem - several of them are P=NP-hard in the general case - and it has to run continuously, in production, faster than a human would ever attempt it by hand. Get it right and clusters run at half the cost. Get it wrong once and you've taken down something live.

This posting covers multiple teams working on different pieces of the same automation problem node placement algorithms and provisioning, deep integrations with tools like Karpenter, workload rightsizing (vertical and horizontal), moving running workloads between nodes without interrupting them, storage that scales itself, and GPU capacity optimization across clouds and regions. All of it runs in real time, in production, with no human tuning YAML in the loop. This is a location-specific opportunity.

We are currently accepting applications from candidates residing in the following European countries: Bulgaria, Croatia, Estonia, Greece, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, Cypru

Descripción completa, empresa y enlace para postular: después de iniciar sesión. Iniciar sesión
Los contactos del empleador (correo/teléfono/telegram) están ocultos en la vista previa pública.
¿Pregunta urgente? Escribe a @termdocs