How AI Fixes Medical Record Errors | $125B Healthcare Problem Solved

Medical documentation errors cost U.S. hospitals over $70 billion in denied claims and $55 billion in lawsuits every year. In this episode, we sit down with Dimitri, Founder & CEO of WorkDone Health, a Y Combinator-backed startup that’s building the "Grammarly for medical records."

WorkDone Health automates chart review, compliance checks, and billing validation in real-time, preventing errors before they cost hospitals money—or compromise patient safety. We explore:

Why CFOs are the first to feel the pain of documentation errors

How AI-powered compliance and quality checks reduce denials

Lessons from Y Combinator and scaling a healthcare startup

Why WorkDone could be the antidote to insurance AI denials

If you’re a healthcare leader, investor, or builder in healthtech, this episode shows the future of clinical documentation.

Timestamps:

0:00 – Intro: The cost of documentation errors ($70B in denials, $55B lawsuits)

1:00 – Dimitri’s journey: From physics to healthcare AI

3:00 – The problem: Reactive vs. proactive documentation review

5:15 – Real-world example: Left vs. right shoulder conflict

7:00 – Sepsis bundle case study

9:00 – How WorkDone Health prevents denials in real time

12:00 – Impact on CFO metrics: denials, lawsuits, billing cycle

14:30 – The “antidote” to insurance AI claim denials

18:00 – How the tool works: real-time vs. batch checks

22:00 – Prioritization of alerts: reducing physician burden

27:00 – Vision: “Grammarly for medical documentation”

29:00 – Lessons from Y Combinator for healthtech startups

32:00 – HIPAA compliance and why it matters from Day 1

37:00 – Future of WorkDone: API integrations with EMRs

Dimtry Karpov: https://www.linkedin.com/in/dmitrykarpov/

WorkDone: https://www.linkedin.com/company/workdonehealth/

WorkDone: https://www.wrkdn.com/

Roupen Odabashian: https://www.linkedin.com/in/roupen-odabashian-md-frcpc-abim-183aaa142/