You don’t have to tell us what to automate.We find it first.
GeminAI Dx spends a few weeks learning how your organization’s work actually happens — every hand-off, every workaround, every repeated click — before a single agent gets built. Then we automate what’s actually worth automating.
From raw activity to running agents.
A short discovery period turns clicks and app usage into a workflow map. GeminAI Dx then builds agents for the workflows worth automating — operational and clinical, always with a person able to step in.
We learn how your organization actually works.
For two to four weeks, our system observes how work moves across your organization — the tools used, the steps taken, and the handoffs between them. It reconstructs recurring workflows automatically and identifies where time is being lost, work is repeated, or processes break down.
- 1Referral received
- 2Eligibility checked
- 3Authorization submitted
- 4Appointment scheduled
We don’t just find the work. We automate it.
Using the context learned in the first phase, GeminAI Dx builds agents directly around your workflows and existing tech stack. Because the system already understands how work gets done, agents can be configured and deployed faster — executing work in the background and escalating only when human judgment is needed.
Two founders, two vantage points on the same problem.
AI product leader and MBA at Harvard Business School, with experience building technology products.
LinkedIn
Physician with years of experience at the intersection of healthcare, data and AI.
LinkedInPhD in Graph Neural Networks (UPC), previously applying machine learning to healthcare data as a postdoctoral researcher at Hospital Clínic Barcelona.
LinkedInMedical oncologist at Massachusetts General Hospital and instructor at Harvard Medical School, researching immunotherapy toxicity and biomarkers.
LinkedInFormer Chief AI Officer at Keralty, with extensive leadership experience across healthcare operations, digital transformation, and AI at scale.
LinkedInCurious what a diagnostic would find in your organization?
See where your teams spend time, which workflows repeat, and what could be automated.
Request a diagnostic