Tech Blog
Contact Us

Introduction

A patient arrives in the ED with a bleeding scalp laceration. Abridge surfaces a perfect match for a clinical trial evaluating machine-learning-based methods to assess bleeding. On paper, it's correct. In practice, you don't ask a bleeding patient in a chaotic ED if they'd like to join a study on measuring bleeding.

The gap between what looks right on paper and what's right for the patient is why I believe the most valuable person in healthcare AI right now is someone who's actually practiced medicine.

I was an interventional cardiologist before joining Abridge. My title is “clinician scientist,” and I’m far from the only one at the company. The truth is, you cannot build good healthcare AI that truly serves the needs of clinicians, health systems, and their patients without clinical expertise in the room. The judgment clinicians develop over years of training and patient care is now a sought-after skill at Abridge.

All of this has given rise to the “clinician-builder,” and it’s our job to ensure we are building for clinicians. The days when technology was forced upon us have come to an end; clinicians are now in the driver’s seat.

This is the first in a series of articles written specifically for aspiring clinician builders. In this article, I will explain what the job actually involves, what I love about it (and what I miss about being a practicing clinician), and how to get involved in building the next generation of clinician-centered technologies.

Why Clinicians Are Essential in the Healthcare Build Loop

It’s worth asking the question: Why do clinicians belong on AI build teams at all? Essentially: judgment.

AI can't invent new healthcare. Healthcare is too complex, too multi-factor, too multi-agent, and too human.

Take our example above, a clinical trial matching system that easily makes a technical match but doesn’t understand the human element. The match is correct on paper but the judgment on how and when to assess the patient needs to be human, and that human needs to be an experienced clinician.

An engineer sees clean eligibility criteria; a clinician sees a chaotic ED and a patient in no state to consent. Closing that gap within the technology is the clinician-builder's job, and it's a gap that matters for patient outcomes.

What the Day-to-Day as a Clinician Builder Actually Looks Like

I started in this new role just a few months ago, and I still haven’t yet figured out what a typical week is.

I went from onboarding straight into prototyping and shipped multiple products ahead of Abridge's Keynote event in New York. At the event, we unveiled our newest capabilities to health system leaders. One of the projects I worked on was a system for interfacing with Abridge that is entirely voice-controlled. In addition to working on the product, I helped film a video using it and explaining how it works. I have also been working on clinical trial matching at the point of care along with several other new parts of the product. And nearly all day every day, colleagues ask me to evaluate their projects.

The work is interesting, challenging, and varied. At the same time, I do miss interventional work, like stents and valves. But, to me, it’s worth the impact I feel I can have. I used to help one patient at a time, and now the work I do could impact millions of patients because this year alone the Abridge clinical intelligence platform will be used in 100 million encounters across more than 300 of the largest health systems in the U.S.

Why I Made the Jump to a Clinician-Led AI Company

When I was practicing medicine, I was always thinking about my administrative frustrations—repetitive paperwork, prescriptions, whatever bureaucratic mess the job created. Once AI came along and made it easier to code, I decided to just start building software to solve these problems.

Then, earlier this year, I tried something I had never tried before: I participated in a hackathon—and I ended up winning! Soon after, I received a message from Dr. Shiv Rao, Abridge's CEO and Co-Founder. He offered a note of congratulations and it was the start of a more meaningful conversation. In addition to leading Abridge, Dr. Rao also still practices as a cardiologist at UPMC in Pittsburgh. We spoke a common language, and when we met in person, the fit seemed obvious. One important thing for me is that Abridge is clinician-led, meaning that my contributions would be valued.

It was also important to me that Abridge pairs clinicians with engineering, security, and privacy experts. Clinicians are trained to take responsibility for clinical decisions, not for security architecture, and healthcare technology demands both. I know I'm safe to be as creative as I want and that the products I help build will be implemented safely.

It also means that the “why” of what we build is informed by the context of practicing medicine. Technology built by clinicians is designed to meet us where we work and not force us to change how we deliver care.

Lastly, at a clinician-led company, the mission really resonates with me and my colleagues. We're trying to bring joy back to practicing medicine—and we know firsthand what’s at stake.

How to Get Started in Healthcare AI

The skills that matter most at Abridge are ones we clinicians already have: creativity, clear communication, and persistence.

You already have the ability and you don't need to change careers to start. Go to health tech conferences, hackathons, and talk to the people building the technologies you use every day. You don't need a title change to start showing up in these spaces. You just need curiosity and a willingness to dig into problems that need solving.

The nearest path runs through your own institution: Raise your hand when your organization pilots AI and become the clinician voice in the room when technology decisions get made. That's the exact muscle clinician-builders use every day, and it's how you can learn whether this work energizes you.

Share

In this series:

PT.01 · CurrentWhat It Takes to Be a Clinician-Builder at AbridgeDr. Michal Nedoszytko
PT.02 · Up nextWhat a Clinician-Builder at Abridge Actually Builds: Inside Clinical Decision SupportDr. Chase Yarbrough

Recent posts

Clinical conversations are full of specialized language, multiple speakers, and noisy environments — and transcription errors propagate through the whole pipeline. How we built an LLM-based ASR model, tuned with reinforcement learning and contextual biasing, to recognize medical terminology more reliably…

No single model is the right tool for every clinical task, which is why we use frontier models and specialized in-house models together. Why we build in-house clinical models, why it’s hard, and how we approach the work — so clinicians never have to think about which model is behind their tools…

Beyond offline evaluation and online monitoring, “in-line” evaluations run inside the note generation pipeline itself. How our Editors detect and correct issues before a clinician ever sees the draft — and how we validate the Editors themselves…

User feedback carries signal of what clinicians care about in real encounters — but it arrives as unstructured comments and note edits. How we decompose that feedback into atomic units and classify it against an error taxonomy of over 50 categories…

Our online evaluations run continually on our deployed note generation systems. Every day we evaluate thousands of notes across many quality dimensions — work that relies on LLM judges, and on validating those judges against expert clinician judgment…

All new features and product improvements at Abridge undergo “offline” pre-deployment evaluation. These evaluations are conducted on systems that are not yet deployed with our partners, enabling experimentation without risking adverse impacts…

Note generation was the first product we built at Abridge, the first product with which we earned the trust of our users, and the first product on which we honed our evaluation framework. Here we show how we answer the question of what makes a great clinical note…

Reliable evaluation is a baseline requirement for developing AI products that users can trust. At Abridge, where our products are trusted by tens of thousands of clinicians, we honor and maintain that trust through rigorous evaluation of those products…

A few months ago, I sat down for a filmed roundtable with another Abridge clinician builder and two product leaders. The topic was Clinical Decision Support, a new product that surfaces evidence-based information during the patient encounter…

A patient arrives in the ED with a bleeding scalp laceration. Abridge surfaces a perfect match for a clinical trial evaluating machine-learning-based methods to assess bleeding — on paper, correct; in practice, wrong…

The question isn’t whether clinicians will use AI for clinical guidance. It’s whether that AI has been rigorously evaluated for enterprise readiness…