Lead Platform Engineer/Architect - HPC, Kubernetes

EPAM Systems; United States; On-site; USD 150,000 - 230,000; Full time

Job summary

EPAM Systems Inc. is building and operating Kubernetes platforms at massive scale across multiple clouds to power thousands of GPUs. You will design, implement, and operate scalable infrastructure, tooling, and job scheduling to maximize GPU utilization and reliability.

We value expert-level Kubernetes, Python and IaC skills, with experience in AWS, GCP, and Terraform. You will collaborate with Networking, Storage, Security, and AI/ML teams to deliver reliable, scalable cloud infrastructure.

Qualifications

• 10+ years of experience in infrastructure engineering, cloud platforms, or HPC environments.
• Expert-level Kubernetes experience at meaningful scale, including node pool sizing, scheduler debugging, CNI troubleshooting, and rolling upgrades across large fleets.
• Advanced Python skills with production-grade tooling; experience with Go, Rust, or C++ is a strong plus.
• Daily proficiency in Terraform for writing and reviewing infrastructure as code.
• Working knowledge of core AWS services including EC2, S3, EFS, and FSx for Lustre.
• Strong site reliability engineering background with monitoring, alerting, and incident response practices.
• CKA, CKS certificates are highly preferred.

Responsibilities

• Operate and scale Kubernetes platforms (EKS, GKE, and other distributions) including cluster lifecycle management, node pool optimization, and networking policies during periods of rapid growth
• Provision and manage HPC infrastructure through CI/CD pipelines spanning AWS, CoreWeave, GCP, OCI, and additional cloud providers
• Design and maintain job scheduling systems that efficiently allocate GPU compute resources across training and inference workloads
• Define SLIs/SLOs, build robust monitoring and alerting systems, and actively participate in incident response and post-incident reviews
• Develop production-quality tooling and automation to support multi-cloud infrastructure operations at scale
• Collaborate daily with Networking, Storage, Security, and AI/ML platform teams to ensure seamless cross-functional infrastructure delivery

Skills

Kubernetes

Python

Go

Rust

C++

Terraform

AWS

GCP

CI/CD

Linux

Tools

EKS

GKE

CI/CD pipelines

Monitoring tooling

Job description

Join a high-growth infrastructure team operating Kubernetes platforms across multiple cloud providers at massive scale. You'll build the systems that power thousands of GPUs, where your code and configurations directly protect thousands of GPU-hours from costly failures. EPAM is where tech talent thrives building groundbreaking solutions, advancing your skills through world-class learning platforms, and working alongside a global community of problem-solvers to make the future real.

Req# 1103823066

Responsibilities

• Operate and scale Kubernetes platforms (EKS, GKE, and other distributions) including cluster lifecycle management, node pool optimization, and networking policies during periods of rapid growth
• Provision and manage HPC infrastructure through CI/CD pipelines spanning AWS, CoreWeave, GCP, OCI, and additional cloud providers
• Design and maintain job scheduling systems that efficiently allocate GPU compute resources across training and inference workloads
• Define SLIs/SLOs, build robust monitoring and alerting systems, and actively participate in incident response and post-incident reviews
• Develop production-quality tooling and automation to support multi-cloud infrastructure operations at scale
• Collaborate daily with Networking, Storage, Security, and AI/ML platform teams to ensure seamless cross-functional infrastructure delivery

Requirements

• 10+ years of experience in infrastructure engineering, cloud platforms, or high-performance computing environments
• Expert-level Kubernetes experience at meaningful scale, including node pool sizing, scheduler debugging, CNI troubleshooting, and rolling upgrades across large fleets
• Advanced Python skills with a track record of building production-grade tools, not just scripts; experience with Go, Rust, or C++ is a strong plus
• Daily proficiency in Terraform for writing and reviewing infrastructure as code
• Working knowledge of core AWS services including EC2, S3, EFS, and FSx for Lustre
• Strong site reliability engineering background with experience building monitoring, alerting, and incident response practices
• CKA, CKS certificates are highly preferred