Lead VLM Pretraining Scientist for Medical Imaging

Epsilon Health; San Francisco (CA); On-site; USD 180,000 - 250,000; Full time

Job summary

Epsilon Health is seeking a Research Scientist to lead pretraining of large vision-language models for radiology report generation. You will work on scale-ready VLM architectures, multimodal data mixtures, and grounding in medical images like X-rays, CTs, and MRIs.

The role sits within the ML Research team and collaborates with post-training partners. The ideal candidate has 6+ years in vision-language modeling, hands-on experience with PyTorch or JAX, and a track record of adapting research to

Qualifications

• 6+ years of academia/industry experience in vision-language modeling or related fields.
• Deep expertise in pretraining large vision-language models and multimodal learning.
• Strong PyTorch or JAX proficiency with multi-GPU/distributed training experience.

Responsibilities

• Own the pretraining stage of radiology-focused VLMs, including architecture design and data mixtures.
• Design and scale vision-language models for medical imaging applications.
• Coordinate dataset curation, distributed training, and checkpoint handoffs to post-training.

Job description

Epsilon Health is seeking a Research Scientist to lead pretraining of large vision-language models for radiology report generation. You will work on scale-ready VLM architectures, multimodal data mixtures, and grounding in medical images like X-rays, CTs, and MRIs.

The role sits within the ML Research team and collaborates with post-training partners. The ideal candidate has 6+ years in vision-language modeling, hands-on experience with PyTorch or JAX, and a track record of adapting research to