Senior ML Infrastructure Engineer – Medical Imaging

Epsilon; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Epsilon Health in San Francisco is seeking an ML infrastructure engineer to design and build the core systems for scalable training of large models in medical imaging research. You will enable fast experimentation and reliable training, collaborating with researchers to ship production-ready code from experimentation to deployment.

Join a mission-driven startup to own distributed training, RL infrastructure, and inference pipelines, ensuring models move smoothly from research to clinical impact

Qualifications

• 6+ years designing, building and operating large-scale distributed systems.
• 2+ years building ML infrastructure in production.
• Strong Python and PyTorch/JAX expertise.
• Experience with research workflows and tooling for ML.
• Experience building infrastructure for ML research environments.
• Deep experience with Kubernetes and cloud infrastructure at scale.
• Distributed training at scale (FSDP, DeepSpeed, Megatron).
• Experience as a technical lead or mentor to engineers.

Responsibilities

• Partner with researchers to understand workflows and future needs.
• Build distributed training infra for foundation models on medical imaging.
• Create high-throughput data loading and preprocessing to saturate GPUs.
• Prototype ideas and deliver production-ready code end-to-end.
• Contribute to production serving and deployment pipelines with backend team.
• Develop RL training stack for online, multi-reward RL at scale.

Skills

Distributed systems

ML infrastructure

Python

PyTorch/JAX

Research tooling

MLOps

Leadership

Large-scale training

Tools

Kubernetes

Cloud infrastructure

DeepSpeed

Megatron

TensorRT

Job description

Epsilon Health in San Francisco is seeking an ML infrastructure engineer to design and build the core systems for scalable training of large models in medical imaging research. You will enable fast experimentation and reliable training, collaborating with researchers to ship production-ready code from experimentation to deployment.

Join a mission-driven startup to own distributed training, RL infrastructure, and inference pipelines, ensuring models move smoothly from research to clinical impact