Hybrid AI & HPC ML Engineer

Hewlett Packard Enterprise Company; Springs (NY); Hybrid; USD 121,000 - 277,000; Full time

Benefits offered by this job

Health & Wellbeing

Professional development

Unconditional inclusion

Job summary

Hewlett Packard Enterprise Company is seeking an AI and Machine Learning Engineer to advance high-performance computing and AI workloads within a hybrid work model, involving two days per week in an HPE office. The role focuses on deploying scalable AI systems, optimizing transformers, and coordinating with software/hardware partners to enhance performance across DL workloads.

You will contribute to AI/ML benchmarks, containerized environments, and data-centric computing, while communicating

Qualifications

• Master's degree or PhD in Computer Science, Engineering, Information Technology or Systems, or relevant field.
• 5+ years of experience in ML/AI and HPC environments.
• Experience with NCCL, HPL benchmarks and AI workloads.

Responsibilities

• Install and configure complex IT infrastructure components (servers, storage, network).
• Develop scripts and configurations for automating deployment.
• Study and improve performance of Large Language Models on HPE GPU servers.
• Analyze server workloads and optimize DL/ML code and hardware use.
• Document guidance and reports on AI workload and model selection.

Skills

ML/AI experience

HPC

Python/C++

Transformer models

CI/CD

Education

Master's or PhD in CS/Engineering/IT or related field

Tools

Slurm

NCCL

Lustre

Weka I/O

Docker/Kubernetes

Job description

Hewlett Packard Enterprise Company is seeking an AI and Machine Learning Engineer to advance high-performance computing and AI workloads within a hybrid work model, involving two days per week in an HPE office. The role focuses on deploying scalable AI systems, optimizing transformers, and coordinating with software/hardware partners to enhance performance across DL workloads.

You will contribute to AI/ML benchmarks, containerized environments, and data-centric computing, while communicating