Projects

Tools and systems I've built to solve problems in medical education, research, and healthcare.

Education Tech

Study Optimization System

Python Anki API Data Analysis

Problem

Medical school requires retaining vast amounts of information across multiple subjects simultaneously. Traditional study methods don't adapt to individual learning patterns.

Approach

Built a system that integrates spaced repetition scheduling with performance analytics. Uses actual review timing data to estimate workload and optimize study sessions.

Outcome

Reduced daily study planning time, improved retention metrics, and enabled data-driven adjustments to learning strategy.

Health Economics

Cost-Effectiveness Analysis Framework

R Health Economics Markov Models

Problem

Evaluating healthcare interventions requires complex modeling of costs, outcomes, and uncertainties. Existing tools often require significant setup time for each analysis.

Approach

Developed a reusable framework for Markov modeling and sensitivity analysis. Includes templates for common analysis patterns and automated visualization of results.

Outcome

Streamlined the CEA workflow, enabling faster iteration on model assumptions and clearer communication of results to stakeholders.

Healthcare Tools

Clinical Decision Support Tools

TypeScript AI Integration Clinical Workflows

Problem

Clinical workflows often involve repetitive information synthesis from multiple sources. Manual processes are time-consuming and prone to inconsistency.

Approach

Created focused tools that automate specific clinical documentation and decision support tasks. Emphasis on fitting into existing workflows rather than replacing them.

Outcome

Reduced documentation burden while maintaining accuracy. Tools designed to augment clinical judgment, not replace it.

Building Philosophy

Start with friction

Every project begins with a real problem I've encountered. If I haven't felt the pain, I don't build the solution.

Ship fast, iterate

Get a working version quickly, then improve based on actual use. Perfect is the enemy of useful.

Fit existing workflows

Tools should integrate into how people already work, not require them to change everything.

Augment, don't replace

Especially in healthcare—technology should enhance human judgment, not substitute for it.