HPC (High-Performance Computing) Consultant

Hybrid
Posted Oct 14, 2026
1 position

Opens LinkedIn

Employment type: Full-time
Experience level: Lead · 10+ years
Posting language: English
Working hours: 40 hours per week
Seniority: Executive
Application method: Apply on LinkedIn (LinkedIn sign-in required)

Job summary

Design, deploy, operate, and optimize large-scale, multi-cloud Kubernetes platforms supporting GPU-enabled AI/ML and HPC workloads. Automate infrastructure with Terraform and Python, troubleshoot platform issues, and support monitoring, incident response, capacity planning, and reliability improvements.

Job details

Work Type: Remote-first, with occasional on-site visits to customer offices. Candidates located near any Meta office location will be preferred. Kindly focus on the candidates from top-tier organizations. As this role focuses on large-scale Kubernetes platforms, we are prioritizing candidates with experience in large enterprise environments. please prioritize candidates with strong hands-on Kubernetes experience at scale, including cluster lifecycle management, node pool sizing, scheduler troubleshooting, networking/CNI issues, upgrades, and production operations in large enterprise environments. Candidates should have a solid understanding of HPC workload scheduling concepts such as gang scheduling, fair scheduling, resource allocation, and Slurm, along with proven Python coding and problem-solving skills demonstrated through recent hands-on development work. Additionally, candidates must be able to confidently discuss their resume projects in depth, including architecture decisions, troubleshooting scenarios, production incidents, and their individual contributions. Please ensure candidates are independently capable of answering technical questions without relying on AI tools during interviews, as strong technical depth, reasoning ability, and real-world operational experience are critical for this role. We are looking for individual contributor, I’ve seen resumes with Architect role background, please who working as Lead or Architect not going to suit for this role. So, look for candidates who can do Python coding and strong experience with Kubernetes, Terraform, Hpc. Interview Process: 1st Interview – Tech Mahindra: Kubernetes, HPC Technical Interview and Python Coding2nd Interview – Amazon: Kubernetes, HPC Technical Interview and Python Coding3rd Interview – Meta: Kubernetes, HPC Technical Interview and Python Coding Submission Format: Candidate Full Name as per passportEmailPhone NumberCurrent LocationWork AuthorizationTotal ExperienceJoining AvailabilitySalary / RateAny relatives working in Tech M (If yes, please specify relation)Ex-Tech M (If yes, year of leaving Tech M) SkillsYears experienceLarge-scale production KubernetesKubernetes troubleshooting — scheduler/CNI/nodes/upgradesGPU/HPC infrastructureTerraform / IaCPython automation (coding)SRE / reliability / incident managementAWS infrastructureAI InfrastructureNVIDIA GPU PlatformCalico/CiliumKarpenterLarge-scale clusters (500+ nodes)Multi-cloud experience (AWS + GCP + CoreWeave) Must Have: Kubernetes + Slurm + NVIDIA GPU + AI/ML Infrastructure + Terraform + Python + AWS (EKS/FSx Lustre) + Monitoring/SRE We are looking for a individual contributor with expertise across Kubernetes, cloud infrastructure, HPC platforms, GPU computing, Terraform, and automation. Depending on experience, candidates may be considered for Platform Engineer, Kubernetes Engineer, HPC Engineer, Cloud Infrastructure Engineer, DevOps Engineer, AI Infrastructure Engineer, or Site Reliability Engineer (SRE) roles. Key Responsibilities Design, deploy, operate, and support large-scale Kubernetes platforms across AWS, GCP, CoreWeave, OCI, and other cloud environments.Manage Kubernetes cluster lifecycle activities including provisioning, scaling, node pool management, upgrades, troubleshooting, and performance optimization.Support AI/ML and HPC workloads, including GPU-enabled compute infrastructure for training and inference environments.Provision and automate cloud and infrastructure resources using Terraform and CI/CD pipelines.Troubleshoot Kubernetes scheduling, networking, storage, and platform reliability issues.Implement monitoring, observability, alerting, SLIs/SLOs, and incident response processes.Collaborate with Networking, Security, Storage, AI/ML, Data Engineering, and Application teams.Develop automation and operational tooling using Python and cloud-native technologies.Participate in production support, root cause analysis, capacity planning, and platform optimization initiatives. Required Skills Kubernetes & Container Platforms Kubernetes (EKS, GKE, AKS, OpenShift, CoreWeave CKS)Cluster Lifecycle ManagementNode Pool ManagementScheduler TroubleshootingCNI TroubleshootingNetworking PoliciesRBACHelmAutoscalingRolling Upgrades Cloud Infrastructure AWS (EC2, S3, IAM, VPC, EKS, EFS, FSx for Lustre)Google Cloud Platform (GCP)OCI (Oracle Cloud Infrastructure)Azure (Preferred)Multi-Cloud Infrastructure Infrastructure as Code & Automation TerraformInfrastructure as Code (IaC)CI/CD PipelinesGitHub Actions / Jenkins / GitLab CIAnsible (Preferred) Programming & Scripting PythonBash/Shell ScriptingAutomation DevelopmentREST API Integration HPC & GPU Infrastructure (Preferred) High-Performance Computing (HPC)SlurmNVIDIA GPU PlatformsGPU SchedulingCUDADistributed ComputingAI/ML Infrastructure Monitoring & Reliability PrometheusGrafanaDatadogSplunkCloud MonitoringSLI/SLO ManagementIncident ResponseRoot Cause Analysis (RCA) Preferred Experience Experience with any of the following is highly desirable: AI/ML Infrastructure PlatformsKubeflowKServeRayMLflowvLLMVector Databases PLAINTEXT

What you’ll do

Requirements

Candidates need hands-on experience operating Kubernetes at enterprise scale, including cluster lifecycle management, scheduling, networking, upgrades, and production troubleshooting, along with HPC/Slurm and NVIDIA GPU infrastructure knowledge. Strong Python development, Terraform, AWS experience (including EKS and FSx for Lustre), and the ability to independently explain technical decisions and production incidents are required.