Machine Learning Infrastructure Engineer

True Talent Solutions Partners; United States; Remote; USD 200,000 - 400,000; Full time

Job summary

True Talent Solutions Partners is seeking an engineer to design, deploy, and maintain distributed ML training and inference clusters for AI-for-science applications. You will build scalable pipelines for petabyte-scale datasets and model training, optimizing parallelization and GPU performance across scales.

Ideal candidates have 2–10 years of experience in large-scale ML infrastructure, with cloud experience (GCP/AWS/Azure) and container orchestration (Docker, Kubernetes).

Qualifications

• 2 to 10 years of experience building large-scale machine learning infrastructure for pretrained or foundation models, including compute, data, inference, and distributed training systems.
• Experience optimizing large-scale training, data, or inference workloads, including low-level GPU performance work with CUDA and JAX.
• Proficiency with distributed training frameworks such as FSDP or DeepSpeed.
• Experience with cloud platforms such as GCP, AWS, or Azure, and with containers and orchestration tools such as Docker and Kubernetes.
• Familiarity with scalable model serving, distributed task management, monitoring, logging, observability, and version control practices.
• Background in physical AI or AI for science, such as robotics, autonomous systems, or biology, and an interest in working across the machine learning lifecycle.
• Comfort working in a fast-paced environment and learning unfamiliar domains.

Responsibilities

• Design, deploy, and maintain distributed machine learning training and inference clusters.
• Build scalable end-to-end pipelines for petabyte-scale datasets and model training across the machine learning lifecycle.
• Evaluate parallelization methods and numerical precision trade-offs at different model scales.
• Profile and debug low-level GPU operations to improve performance.
• Track relevant research and apply promising ideas to infrastructure and model workflows.

Skills

Distributed ML infra

GPU optimization

CUDA

JAX

FSDP/DeepSpeed

Cloud platforms

Tools

Docker

Kubernetes

CUDA

JAX

Job description

About the Role

Join an AI for science team building large-scale models for physical systems. You will develop the infrastructure behind distributed training and inference, helping the team advance models that can predict outcomes and inform effective actions.

What You’ll Do

• Design, deploy, and maintain distributed machine learning training and inference clusters.

• Build scalable end-to-end pipelines for petabyte-scale datasets and model training across the machine learning lifecycle.

• Evaluate parallelization methods and numerical precision trade-offs at different model scales.

• Profile and debug low-level GPU operations to improve performance.

• Track relevant research and apply promising ideas to infrastructure and model workflows.

What We’re Looking For

• 2 to 10 years of experience building large-scale machine learning infrastructure for pretrained or foundation models, including compute, data, inference, and distributed training systems.

• Experience optimizing large-scale training, data, or inference workloads, including low-level GPU performance work with CUDA and JAX.

• Proficiency with distributed training frameworks such as FSDP or DeepSpeed.

• Experience with cloud platforms such as GCP, AWS, or Azure, and with containers and orchestration tools such as Docker and Kubernetes.

• Familiarity with scalable model serving, distributed task management, monitoring, logging, observability, and version control practices.

• Background in physical AI or AI for science, such as robotics, autonomous systems, or biology, and an interest in working across the machine learning lifecycle.

• Comfort working in a fast-paced environment and learning unfamiliar domains.

Compensation & Benefits

Base salary range: $200,000 to $400,000 annually. Visa sponsorship is not available.

Location

Remote within the United States, with San Francisco, California as the primary location.