Case Study: RNCalc & The Velocity of Care

Project: RNCalc.app
Role: Product Architect (PharmD)
Timeline: 24-Hour Sprint (Feb 14, 2026)

The Clinical Gap

In Oncology and Critical Care, the gap between medical data and bedside application is often measured in years. While platforms like MDCalc exist for physicians, the nursing workflow—which requires rapid, regulated documentation and safety checks—is often underserved.

I built RNCalc to close that gap. My goal was to create a tool that matches the cognitive complexity of the nursing shift, not just a simple calculator.

The Solution: Clinical Logic, Digitized

RNCalc is a comprehensive decision-support engine designed specifically for the nursing workflow. It goes beyond simple math to handle the regulatory and safety constraints of the bedside.

Visit Live Application: RNCalc.app →

Note: The interface prioritizes high-contrast legibility and zero-latency interaction for critical care environments.

The Feature Set (Shipped in v1.0):

• 25+ Curated Calculators: Specific to high-acuity nursing needs, not just general medicine.
• Safety Context: "Adult vs. Pediatric" toggles that instantly re-calibrate safety limits and formulas.
• Regulatory Compliance: 17 pre-made charting templates validated against medical, nursing, and state regulations to ensure "Protected Charting."
• Workflow Integration: 7 "Nursing Brain" sheets designed to structure the chaotic 12-hour shift.
• Clinical Rationale: Every calculator includes the underlying guidelines and primary source citations, ensuring the user understands the why, not just the result.

The Execution: A 24-Hour Sprint

Traditionally, building a suite of 25 validated clinical tools with export capabilities would take a dedicated team 3–6 months. By leveraging AI-assisted development, I compressed this roadmap into 24 hours.

This wasn't about asking a chatbot to "write code." It was about Clinical Product Architecture. I acted as the Lead Architect and Subject Matter Expert, defining the clinical parameters (e.g., "Ensure the pediatric drip rate alerts if >20mcg/kg/min"). I then directed AI agents to handle the syntax, unit testing, and frontend deployment.

GENERATING_DIAGRAM...

1. Product Architect (Daniel Tran): Defined the clinical needs, provided deep pharmacological context, and established the validation models for critical medication safety.
2. Lead Coordinator (Claude Opus): Operated in Task Assignment mode, focusing entirely on task decomposition, assignment, and synthesizing the results from the team.
3. Agent Team (Claude Sonnet): Multiple independent instances working in parallel via a Shared Task List. Teammates handled Pharmacological Logic, Nursing Workflow UI, and Regression Testing simultaneously.
4. Validation Review: Critical clinical logic was implemented using a validation review process, where teammates remained in read-only status until the methodology was verified by Daniel Tran.
5. Infrastructure (Gemini CLI & Cloudflare Pages): Utilized Gemini CLI for deployment coordination to Cloudflare Pages, ensuring the application remains highly available and responsive even on slow hospital networks or "dead-zones."

The Strategic Implication

This project demonstrates a new reality for Biotech and Pharma: We can compress development timelines through structured delegation.

In Oncology, where protocols change rapidly, we no longer need to wait for quarterly software release cycles to update clinical decision support. We can identify a need, validate the clinical logic, and deploy a compliant, secure tool to the edge in days.

RNCalc is proof that when you combine deep domain expertise (PharmD) with AI-assisted development, the constraint is no longer technical feasibility—it's just our imagination.

Authored by Daniel Tran — Product Architect & PharmD.