Research Scientist

About Us

Epsilon Health is transforming medical imaging and diagnostics by bringing frontier AI into clinical practice. We're tackling some of healthcare's most critical challenges, building technology that directly improves patient outcomes.

This year, we've served over 300,000 patients and now serve 3,000+ daily, working with some of the largest medical institutions in the U.S. By year-end we will interpret 1% of all daily U.S. X-rays.

Our research team brings experience from DeepMind and Meta, powered by one of the industry's most comprehensive proprietary medical imaging datasets, continuously enriched by real-world clinical data. Together, we're building a future where frontier AI helps save millions of lives today, not tomorrow.

Recently out of stealth, we're proudly backed by Jack Altman, Uncork Capital, Renegade Partners, and AlleyCorp, one of NYC's most active healthcare venture firms.

 

Role Overview

We're seeking a Research Scientist with deep expertise in large-scale vision-language pretraining to join our ML Research team. You'll be at the forefront of developing state-of-the-art multimodal models for clinical use in radiology settings. This role owns the pretraining stage of our radiology report generation model: VLM architecture design, multimodal data and task mixtures, and the large-scale training runs that build grounded visual understanding across X-rays, CT scans, and MRI. You'll work with one of the largest and most diverse medical imaging datasets in the industry, paired with the reports that make multimodal pretraining at this scale possible, while maintaining the clinical rigor required for healthcare deployment. Post-training and RL are owned by a partner role you'll collaborate with closely.

Key Responsibilities

• Design, train, and scale vision-language foundation models for radiology applications, owning the pretraining stage end to end.

• Develop VLM architectures suited to medical imaging, including native and variable resolution handling, high-resolution tiling, connector design, and token budgets for volumetric studies.

• Build and tune multimodal pretraining mixtures across captioning, VQA, grounding, and retrieval tasks, balancing data sources to avoid regressions in language capability.

• Develop fine-grained visual grounding during pretraining, enabling models to localize findings within medical images using bounding boxes or segmentation masks.

• Own pretraining evaluation (zero- and few-shot transfer, probing, and downstream fine-tunability) — as the signal for base model quality.