All work BerryWell Institute

From a founder’s idea to a live training platform.

Dr. Eva Berry, DNP, CRNA, wanted anesthesia professionals to practice clinical judgment before the stakes are real. We designed and built the platform that makes it possible: the learner app, the dashboard for her team, the cloud system behind them, and the website.

  • 2026
  • Learner application
  • Admin Center
  • Cloud infrastructure on AWS
  • Analytics and insights
  • Public website
Visit the live site Step through a case
BerryWell Institute website
The live site, as visitors see it today.

At a glance

1,200+ Automated checks before every update
3 Stages every change passes through before it goes live
2 Apps on one platform: one for learners, one for the Institute’s team
Oct 2026 Open to the public, with free registration
The challenge

Clinical judgment usually takes years of practice, one patient at a time. Dr. Berry wanted learners to build it sooner: work through a case as it changes, decide what to do, see what happens, and reflect on it with a mentor. That needed more than a website. It needed a platform that runs complex cases smoothly, keeps the Institute’s content private, shows the team how every learner is doing, and belongs to the Institute.

What we built

Every piece, start to finish.

  1. 01

    A workspace built around the decision

    Learners see the patient, the scene, and the monitor side by side, choose what to do, and watch the case respond. Every case ends with guided reflection and a short assessment, and can be replayed to try a different path.

  2. 02

    A monitor that reacts

    Vital signs, alarms, and ventilator settings change with every decision. They are drawn live on screen, in step with each scene, rather than recorded into a video.

  3. 03

    A dashboard for the Institute

    A separate app where the team reviews results, follows each learner’s progress, and chooses which cases are open to the public.

  4. 04

    Private by design

    Every account signs in securely and sees only what it should. Videos are protected, and each published case is locked to its version, so it never changes in the middle of someone’s practice.

  5. 05

    Built to stay reliable

    Every update passes more than 1,200 automated checks and moves through testing and staging before it goes live. The whole system is documented in code, so it can be reviewed, rebuilt, and handed over.

  6. 06

    The website that introduces it

    We also built berrywellinstitute.com. It explains the experience without giving it away, and every inquiry is safely queued, so no request is lost even if email goes down.

The result

BerryWell Institute opened to the public on October 1, 2026. Healthcare learners and professionals can register for free and try the demonstration cases, and the Institute runs the platform from its own dashboard, on infrastructure it owns.

See it live at berrywellinstitute.com
How a case unfolds

Step through it.

Healthcare learners and professionals can try the Institute’s demonstration cases for free. Tap each step to see what the learner does, and what the platform does behind it.

An illustration of the layout drawn by AI Prodigy, not a screenshot.

  1. Review the patient’s context, goals, and starting conditions.

    Platform The brief stays on screen through every scene.

  2. Read vital signs, waveforms, alarms, and ventilator settings as the situation changes.

    Platform Signals are drawn live from the case, not played from a video.

  3. Choose among clinically plausible actions.

    Platform The choice is saved, and the case branches.

  4. Watch the patient and the situation respond, then reassess.

    Platform The next scene and the monitor follow the path the learner took.

  5. Reflect with the mentor, then answer targeted assessment questions.

    Platform Results reach the Institute’s dashboard, and the learner can replay to try another path.

Under the hood How the platform is built, for technical readers
Applications
  • Public website Prerendered pages and inquiry routing
  • Learner application Brief, scenes, monitor, ventilator, and notes
  • Admin Center Results, insights, and scenario availability
Services
  • Sign-in and permissions Amazon Cognito, role-based access
  • Platform API Amazon API Gateway and AWS Lambda
  • Operations CloudWatch alarms, scheduled jobs, and cost budgets
Data
  • Learner and program data PostgreSQL on Amazon RDS
  • Protected media Amazon S3 behind CloudFront signed links
  • Website inquiries Queued, archived, and delivered by email

Defined in code with the AWS CDK, released through GitHub Actions, and run in the Institute’s own AWS account.

Have a project like this in mind?

Or something smaller. Tell us what you need, and we’ll show you the simplest way to build it. The first conversation is free.

Get a free audit See other work

The audit and first consultation are free. You pay nothing unless we agree on it in writing.
Prefer email? contact@ai-prodigy.com