AI Agents in Healthcare Ops 101

A hands-on course on what AI agents can actually do for healthcare operations today, taught through real revenue-cycle and network workflows: eligibility, denials, credentialing, referrals, and prior auth.
20% off
for groups of 3 or more
Ask us about group discounts and bundles!
$300 off
per seat for groups of 2 +
Ask us about group discounts and bundles!
$150 off
per seat for groups of 3 +
Ask us about group discounts and bundles!
$300 off
per seat for groups of 2 +
Ask us about group discounts and bundles!
$300 off
per seat for groups of 2+
Ask us about group discounts and bundles!
$200 off
per seat for groups of 2 +
Ask us about group discounts and bundles!
$200 off
per seat for groups of 3 +
Ask us about group discounts and bundles!
$300 off
per seat for groups of 2 +
Ask us about group discounts and bundles!

What this course covers

Healthcare runs on manual operational work: verifying insurance, chasing claims, credentialing providers, processing referrals. This course starts from those pain points and works forward to how production AI agents handle them today, with live demos, a hands-on denial lab, and a capstone where you design a deployment for a workflow you choose. You'll leave able to answer three questions about any workflow: can an agent do this today, what does it take to deploy it safely, and how do I know it's working?
No items found.
No items found.
No items found.

Things You'll Get From This Course

Read the landscape

Understand the core revenue-cycle and network workflows (eligibility, claims follow-up, denials and appeals, credentialing, referrals, prior auth): what they are, why they're painful, and what they cost when they break.
1

Know what agents can and can't do

Build a working mental model of modern AI agents, how they differ from RPA bots and chatbots, and where human judgment still has to stay in the loop
2

See real deployments, not demos

Study how production agents run eligibility verification, denial recovery, and credentialing end-to-end, including the results that matter and what the humans do now
3

Leave with a playbook

Learn the guardrails (HIPAA, human-in-the-loop checkpoints, audit trails, accuracy and ROI measurement) and build a deployment plan for a workflow in your own organization
4

Meet Your Instructor, Ted Cheng

Alex Dou

Alex Dou is Chief of Staff at Out-Of-Pocket

Before OOP he worked in and around healthcare, mostly on the payer side, as a Product Manager, so he's learning a lot with all these courses. Wow! There's a lot of stuff all happening outside of the walls of the payer! 

He asks a lot of "dumb" audience surrogate questions, which makes him an ideal, if somewhat wordy, moderator for these courses

Register For This CourseSignup Re-Opening Soon
Next Cohort Starts
10/27 - 10/29

Course Syllabus & Schedule

Module 1

Day 1

Eligibility, denials, and the workflows that AI should eat first

(10/25, 12-1:30PM EST)

Today we'll hit 2 things: 1. Which healthcare ops workflows we've found to deserve AI, sorted by pain + volume || 2. Defining AI agents here in 2026 Anno Domini - how they differ from RPA and chatbots, and where their limits are (aka hand off to humans). || We'll teach this with a live demo of an agent running a real eligibility check on a portal and a group exercise mapping a workflow's failure points

Module 2

Day 2

Zoom in on RCM flows + hands-on lab working a denial

(10/26, 12-1:30PM EST)

Today we'll go on a deep dive: 1. Eligibility Verification and 2. Denial Management. || The day ends with 2 exercises: 1) a hands-on lab: work a real de-identified denial with an AI: classify it, appeal vs. resubmit, draft the letter 2) you'll also do a "red-teaming" exercise: you'll explore tricky conditions and setup that will break an agent to learn its limitations

Module 3

Day 3

Zoom back out

(10/27, 12-1:30PM EST)

1. A fast tour of the network-ops long tail: credentialing, referral intake (by fax, yes, still), prior auth, etc. We'll use what we learned in Day 2 to evaluate these workflows || 2. The messy questions that block moving to production: HIPAA, where you should be careful with data handling, what's the implications with model training, and why your personal ChatGPT should not be used for this type of automation || The day will end with a capstone workshop where you build a deployment plan + get feedback from the group

Module 4

Day 4

No items found.

Module 5

Day 5

No items found.

Module 6

Day 6

No items found.

Module 7

Day 7

No items found.

Module 8

Day 8

No items found.

Module 9

Day 9

No items found.

Module 10

Day 10

No items found.

Frequently Asked Questions

Who is this course for?

Operations, revenue-cycle, and product leaders at care-delivery companies who are being asked "should we use AI agents for this?" and want a grounded answer. Also useful for founders, new healthcare hires, investors, and anyone building or buying automation for the back office. No technical background required.

Who is this course NOT for?

If you're looking to build AI agents from scratch (prompt engineering, model selection, writing code), this isn't that course. It's also not a clinical AI course: we stay on the operational side (eligibility, claims, credentialing, referrals), not diagnosis or clinical decision support.

Do I need a technical background?

No. We explain agents from the operations side: what they can do, where they fail, and how to manage them. If you can describe a workflow to a new hire, you have what you need. The hands-on lab uses an AI assistant, not code.

Will I be an AI agents expert after finishing this course?

No, but you'll be a much better judge of one. You'll be able to look at a workflow and say whether an agent can handle it today, what guardrails it needs, and what to ask a vendor before you sign. Deploying and operating agents well takes practice; this course is where the practice starts.

Is there a lot of work?

No homework. The course is three 90-minute sessions, and the hands-on parts (the denial lab and the capstone plan) happen in class. If you want to get more out of the capstone, come with a workflow from your own organization in mind.