Who Is Eduardo Reis
The Doctor Who Went Full Stack
Most doctors pick a lane. Eduardo Pontes Reis, MD picked all of them simultaneously. A trained neuroradiologist from Brazil with a fellowship at Hospital Israelita Albert Einstein - one of Latin America's foremost medical institutions - he arrived at Stanford's Center for Artificial Intelligence in Medicine and Imaging (AIMI) not as a passenger but as a builder. Within a couple of years, he had published datasets in Nature, open-sourced machine learning models, helped spin out a startup, and watched it sell for eight figures. In early 2026, he joined Hippocratic AI as Vice President of Strategy in the Office of the CEO, arriving at a company that has raised $439 million to put patient-facing AI agents into health systems, payer organizations, and life sciences companies worldwide.
What separates Reis from the typical "physician-turned-tech" narrative is the absence of a clean conversion moment. He never left medicine to do AI. He made them occupy the same room, argue with each other, and eventually ship something useful. The BRAX dataset he co-authored at Hospital Israelita Albert Einstein contains 40,967 labeled chest X-ray images - scans from real Brazilian patients - and became a public resource for training AI models in settings where English-language data doesn't translate. The labels were derived using NLP from radiology reports written in Portuguese. The research crossed language barriers, geographic barriers, and the traditional boundary between clinician and engineer.
"Deep expertise at the intersection of clinical medicine, medical imaging, and artificial intelligence."
- Hippocratic AI press release, February 2026, on appointing Eduardo Reis as VP Strategy
That intersection is exactly what Hippocratic AI needs. The company's platform - built around patient-facing conversational AI agents - requires someone who understands both the clinical stakes of a wrong answer and the technical architecture required to prevent one. Reis brings both. At Albert Einstein in Brazil, he led AI development for diseases of national significance: tuberculosis, COVID-19, melanoma, and head CT analysis. He didn't just study these applications - he built them, evaluated them in clinical settings, and thought hard about what happens when they fail.
His open-source footprint tells a parallel story. On GitHub under the handle edreisMD, Reis has 49 repositories and contributions to high-profile projects including LangChain and Open-Assistant. He built plugnplai, a Python library for integrating AI plugins into large language models, before plugin ecosystems became fashionable. He built Comp2Comp, converting CT scans into body composition analysis - a tool with 115 stars and real clinical utility. He earned GitHub's "YOLO" badge, which requires merging a pull request without a code review. A neuroradiologist. Who YOLOs merges. That's the contradiction at the center of Eduardo Reis.
Between Stanford and Hippocratic AI, Reis served as Founding Radiologist at Cognita Imaging, an AI startup that spun out of Stanford's AIMI lab. Cognita developed vision language models for analyzing CT and X-ray images and drafting initial radiology reports - a direct commercial application of the research Reis had been doing for years. The company operated in stealth. Then, in November 2025, Radiology Partners acquired it for approximately $80 million. Three months later, Reis was announced as VP Strategy at Hippocratic AI, joining alongside a new Associate Chief Medical Officer and Chief Marketing Officer as the company scales following its $141 million Series C.
