PUBLIC ENTRY PAGE
Know Thyself
Four AI agents. One dying patient. Built so students make their worst mistakes here — in real emergencies, there are no do-overs.
ROUND 1 DEADLINE
VOTING CLOSES THURSDAY, MARCH 26
Know Thyself is live in Round 1 right now. Voting closes Thursday, March 26, so if you're backing this project, send people into the matchup before the round locks.
Builder
Peter Lacko
Build Type
Agent Team
Lifecycle
Working prototype
Consensus Score
83.1
Region
REGION 3
Seed
10
Current Matchup
REGION 3 Round 1 • Round 1Opponent
Meta-Stamp: PocketsProject Link
https://know-thyself-mvp.vercel.appCATEGORIES
EducationAutomation / WorkflowOther
Go Deeper
Most medical simulations teach students what to do. Know Thyself teaches them the consequences when they don't.
Four AI agents run a complete learning cycle: the Cognitive Coach activates the clinical knowledge before the crisis. The Core Agent becomes both narrator and patient — deteriorating or improving in real-time with every decision you make. The AAR Agent debriefs on what went wrong and why. The Paramedic Master authors the clinical blueprints that make it medically accurate.
The consequences are real. Wrong drug — vitals crash. Missed assessment — patient deteriorates. Hesitation — vitals deteriorate, patient status drops. No hints, no scoring, no interruptions — just outcomes unfolding the way they do on a real call.
Under the hood: 30+ medications parsed from natural language, evidence-based physiological transitions, infinite emergency scenarios. Built solo by a licensed paramedic. Powered by Claude.
This isn't a chatbot wearing a stethoscope. It's the sim where users make every fatal mistake before real people pay for it.
Next: a version for the public, then ER and ICU physicians.
Stack Used
Claude Code, React 19, TypeScript, Node.js/Express, PostgreSQL, Prisma, Vercel, Koyeb