Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

The vibe code shift in healthcare

Solutions move closer to the problems.

[Photo: Getty Images]

Profile photo of Ryan Hungate

Chief Strategy Clinical Officer · Henry Schein One

For two decades, healthcare has followed the same pattern for fixing a broken technology workflow: A clinician or administrator identifies a problem, describes it to an internal or vendor software team, and waits. Months later, sometimes years, a tool arrives. It rarely fits the problem exactly as described, because something inevitably gets lost in translation.

Vibe coding—describing what you want in plain language and letting AI generate a working version—is starting to close that gap, and healthcare may be one of the industries best positioned to benefit.

What vibe coding changes

The term describes a real shift in how software gets made: Instead of writing code line by line, a person describes an outcome in natural language. Then, an AI system translates that description into working software, refining it through conversation rather than syntax.

It’s been adopted quickly in general software development and is now being adopted in regulated, high-stakes industries like healthcare.

In the past, coding ability was the gatekeeper for building software. Understanding a problem and having the technical skill to solve it were two separate jobs, held by two different people. Vibe coding lowers the skill level you need to get started, which means the population of people who can act on their own knowledge of a problem expands dramatically.

That matters more in healthcare than almost anywhere else, because healthcare’s core problem has never been a shortage of people who understand what’s broken.

The knowledge is not the bottleneck

Ask any clinician, office manager, or billing lead what’s actually broken in their day. In my experience, it’s often the call that takes 20 minutes on hold with an insurer, or the twice-rewritten note, once for the chart and once for the payor.

It’s the referral that disappears into a fax machine and resurfaces three weeks later. The people closest to these problems can describe them with total precision. But they’ve lacked the ability or opportunity to act on those problems directly.

The traditional technology solution required knowledge to travel through a long chain: a request, a prioritization meeting, a development queue, a build cycle, then a rollout. The tool usually arrived long after the original problem stopped being anyone’s top concern.

Vibe coding shortens that chain by letting the person who understands the workflow build the first-fix version in the time it used to take to write the request.

That’s a meaningful change in an industry where administrative overhead can contribute to clinician burnout and lost capacity, in my experience. A different group of people is doing the building, and the distance between noticing a problem and acting on it is much shorter.

Built-in risk management

None of this works if it’s treated as a shortcut around discipline. Healthcare runs on data that carries some of the most stringent legal, financial, and criminal consequences of any industry. If it’s mishandled, the consequences are real. A tool that makes it easy for a non-engineer to build something can just as easily make it easy for a non-engineer to build something unsafe if the guardrails aren’t in place before they start.

So, legal and perhaps even regulatory responsibility needs to shift from the individual builder user to the technology provider. A first-time vibe-coding builder shouldn’t be expected to configure permissions, data boundaries, and security correctly. They have to be the default setting, invisible unless something goes wrong.

A scheduling-workflow automation tool carries risks such as failing to adhere to federal discharge policies or leaking Protected Health Information (PHI). Another tool that touches clinical documentation or decision support requires clinician validation. Shared information between technology and clinicians must be safe, secure, and meet healthcare industry standards and regulations.

The industries that get this right will be the ones that make it safe enough to open the door at all.

What this looks like

The shift is already visible in small ways. I’ve seen a billing director prototype the exact claims dashboard her team needs instead of waiting on a vendor’s generic version, and a practice leader build a reporting view that matches how her organization is actually structured, rather than adapting her operations to fit someone else’s software.

Together, these approaches represent how the fix gets built by the person who found the problem, not routed through a queue to someone who has to imagine it secondhand.

Over the next several years, that’s likely to matter more than any single new application. Healthcare needs the people who understand its problems best to finally have a way to act on that understanding directly, without waiting for someone else to get around to it or inheriting new risk in the process.

Ryan Hungate is chief strategy and clinical officer at Henry Schein One.