Cloud DevOps Engineer (GCP)

EPAM Systems Inc; United States; Remote; USD 150,000 - 210,000; Full time

Job summary

EPAM Systems in the United States seeks a Senior/Lead Cloud DevOps Engineer to provision and manage the GCP/GKE infrastructure foundation for TPU workloads, including topology-shaped node pools and cost guardrails.

You will establish quotas, handle TPU capacity planning, and set up logging/monitoring with Prometheus, Grafana, and Cloud Logging, collaborating on networking, security, and IAM configurations. Knowledge of FinOps is a plus.

Qualifications

• Strong GCP and Kubernetes expertise required.
• IaC proficiency with Terraform/Helm essential.
• Scripting skills in Python and Bash.

Responsibilities

• Provision GKE topology-shaped node pools using Terraform/Helm.
• Establish quota and cost guardrails for TPU accelerator usage.
• Manage TPU capacity as a project-critical dependency (quota requests, reservations, regional availability).
• Set up logging/monitoring (Prometheus, Grafana, Cloud Logging).
• Support cloud networking, security, and IAM configuration for TPU workloads.

Skills

GCP

Kubernetes

Terraform

Helm

Python

Bash

Monitoring

IAM

FinOps

Tools

Terraform

Helm

Prometheus

Grafana

Cloud Logging

GKE

Job description

We are looking for a Senior/Lead Cloud DevOps Engineer to provision and manage the GCP/GKE infrastructure foundation for TPU workloads, including topology-shaped node pools, automation, and cost/quota guardrails.

Responsibilities

• Provision GKE topology-shaped node pools using Terraform/Helm
• Establish quota and cost guardrails for TPU accelerator usage
• Manage TPU capacity as a project-critical dependency (quota requests, reservations, regional availability)
• Set up logging/monitoring (Prometheus, Grafana, Cloud Logging)
• Support cloud networking, security, and IAM configuration for TPU workloads

Requirements

• Strong GCP expertise (GCE, GKE) and Kubernetes container orchestration
• Infrastructure-as-Code proficiency (Terraform, Helm)
• Scripting proficiency (Python, Bash)
• Experience with monitoring/observability stacks (Prometheus, Grafana, Cloud Logging)
• Understanding of cloud networking, security, and IAM
• Experience running/fine-tuning LLM models is a plus

Nice to have

• Experience managing accelerator (GPU/TPU) quota and capacity planning at scale
• Cloud financial/cost-optimization experience (FinOps)