Note: The job is a remote job and is open to candidates in USA. KPG99 INC is recruiting for an unnamed client seeking an HPC AI/ML Platform Engineer. The role supports and scales multi-cloud infrastructure for GPU-intensive model training, deep learning research, and production AI workloads, with a focus on Kubernetes, GPU resource allocation, Slurm scheduling, and automation.
Responsibilities
- Engineers on this team will help provision and operate large-scale compute environments, manage Kubernetes-based infrastructure, optimize GPU resource allocation, support Slurm scheduling systems, and build automation that enables rapid AI platform growth
- Provision HPC infrastructure through CI/CD systems across AWS, CoreWeave, GCP, OCI, and additional cloud providers
- Operate and troubleshoot large-scale Kubernetes platforms supporting GPU-intensive workloads
- Define and maintain SLIs/SLOs, monitoring, alerting, incident response, and post-incident reviews
- Partner closely with Networking, Storage, Security, and AI/ML Platform teams
Skills
- USC/GC
- Hands-on experience supporting High Performance Computing (HPC), AI/ML, Deep Learning, or GPU-intensive production environments at enterprise scale
- Candidates should demonstrate experience supporting model training workloads, distributed compute platforms, or large-scale AI infrastructure
- Strong experience operating production Kubernetes platforms supporting high-performance workloads, including cluster lifecycle management, scheduling, networking, upgrades, troubleshooting, and platform reliability
- Experience supporting Slurm or similar workload schedulers used to allocate compute resources across AI model training and inference workloads
- Strong Terraform and Python experience for provisioning, managing, automating, and scaling infrastructure environments
- Proven experience supporting High Performance Computing (HPC) environments at enterprise scale
- Experience supporting AI/ML or Deep Learning infrastructure, including model training workloads
- Provision HPC infrastructure through CI/CD systems across AWS, CoreWeave, GCP, OCI, and additional cloud providers
- Experience managing or supporting Slurm-based job scheduling environments
- Operate and troubleshoot large-scale Kubernetes platforms supporting GPU-intensive workloads
- Define and maintain SLIs/SLOs, monitoring, alerting, incident response, and post-incident reviews
- Experience operating environments where performance, cluster stability, and GPU utilization directly impact business outcomes
- 4+ years supporting HPC, AI/ML Infrastructure, Machine Learning Platforms, or Deep Learning environments
- Strong experience with distributed compute, GPU infrastructure, or large-scale training environments
- Cluster lifecycle management
- Scheduler troubleshooting
- Node pool management
- Networking and CNI troubleshooting
- Production upgrades across large fleets
- Terraform experience building and managing infrastructure through Infrastructure as Code
- Python experience for platform automation and operational tooling
- Working knowledge of AWS services including EC2, S3, EFS, and FSx for Lustre
- Understanding of GPU utilization, workload orchestration, and distributed compute environments is highly desired
- Experience With Slurm Job Scheduling Strongly Preferred
- Experience supporting ML training, inference, deep learning, large language model (LLM), or GPU-intensive workloads preferred
Benefits
- 100% remote
Company Overview