Hybrid AI HPC Infrastructure Engineer (GPU/ML)

Analysis Group, Inc.; Boston (MA); On-site; USD 150,000 - 170,000; Full time

Benefits offered by this job

Discretionary annual bonus

Benefits package

Job summary

Analysis Group, Inc. seeks an experienced AI HPC Infrastructure Engineer to own and operate a hybrid HPC/AI computing environment. You will manage Linux clusters, GPU fleets, and multi-tenant workloads to support researchers and data scientists.

You will optimize performance with MPI/OpenMP, implement scalable GPU training/inference, and build MLOps pipelines. Strong scripting, collaboration, and problem-solving are essential in a fast-paced research setting.

Qualifications

• Bachelor's degree in computer science, electrical engineering, or related field.
• 5+ years hands-on Linux systems administration in research/HPC/production.
• Experience with Posit Workbench (RStudio Server Pro) and Python/R environments.
• Experience with SLURM, Platform LSF, or other job schedulers; GPU scheduling preferred.
• Hands-on NVIDIA GPU stack (CUDA, cuDNN, NCCL) and containerization knowledge.
• Familiarity with ML frameworks (PyTorch, TensorFlow) and distributed training.
• Experience with MLOps tools (MLflow, Kubeflow).
• Proficiency in remote access (SSH/RDP) and GPFS.
• Excellent troubleshooting, documentation, and collaboration skills.

Responsibilities

• Maintain and expand HPC/AI compute environment for researchers and data scientists.
• Optimize performance using MPI/OpenMP and distributed multi-GPU training.
• Design and maintain GPU-accelerated infrastructure for large-scale model training.
• Manage GPU scheduling across SLURM/Kubernetes multi-tenant clusters.
• Administer NVIDIA stack and health monitoring for GPU fleets.
• Tune LLM training/inference (batching, quantization, KV-cache).
• Build MLOps pipelines (MLflow, Kubeflow) for training and deployment.
• Manage containers (Docker, Kubernetes, Singularity) for HPC/ML workloads.
• Develop automation scripts and usage reporting across resources.
• Ensure storage and data pipelines on GPFS are robust and performant.
• Troubleshoot hardware/software/network issues and perform root-cause analysis.
• Participate in 24x7 on-call rotation and provide remote/on-site support.

Education

Bachelor's degree in CS/EE

Tools

Posit Workbench

GPFS

Docker

Kubernetes

Singularity/Apptainer

MLflow

Kubeflow

Ansible

Job description

Analysis Group, Inc. seeks an experienced AI HPC Infrastructure Engineer to own and operate a hybrid HPC/AI computing environment. You will manage Linux clusters, GPU fleets, and multi-tenant workloads to support researchers and data scientists.

You will optimize performance with MPI/OpenMP, implement scalable GPU training/inference, and build MLOps pipelines. Strong scripting, collaboration, and problem-solving are essential in a fast-paced research setting.