ML Infra Engineer: GPU Fleet & Orchestration

Generalist AI; San Mateo, Somerville (CA, MA); On-site; USD 180,000 - 240,000; Full time

Job summary

Generalist is building general intelligence for the physical world, operating large-scale GPU infrastructure and on-prem hardware for distributed training and robotics inference. You will own and optimize GPU fleets and data pipelines to empower researchers across workloads.

You will contribute deep expertise in ML hardware, storage, and networking, leveraging Slurm and Kubernetes to orchestrate ML tasks and robot inference fleets in distributed environments.

Qualifications

• Own and manage GPU compute fleets for research and training.
• Ensure GPUs are accessible and maximally utilized for researchers.
• Build and optimize data loading and storage for distributed workloads.
• Experience with GPU orchestration and ML infrastructure.

Responsibilities

• Owning our GPU compute fleets
• Ensure GPUs are easy for researchers to use and maximally utilized
• Optimizing and improving ML data loading transport and storage in highly distributed fully utilized environments.
• Orchestration of robot inference fleets

Skills

GPU fleet management

Kubernetes for ML

Slurm for ML

Large-scale distributed training

ML data loading optimization

NVidia GPU ecosystem

Tools

Slurm

Kubernetes

NVIDIA CUDA

Job description

Generalist is building general intelligence for the physical world, operating large-scale GPU infrastructure and on-prem hardware for distributed training and robotics inference. You will own and optimize GPU fleets and data pipelines to empower researchers across workloads.

You will contribute deep expertise in ML hardware, storage, and networking, leveraging Slurm and Kubernetes to orchestrate ML tasks and robot inference fleets in distributed environments.