I’m a full-stack developer specializing in digital healthcare. With dual majors in optometry/opticianry and computer engineering, I bring interdisciplinary expertise across both fields.
I work end-to-end—from planning and UX/design to development, deployment, and operations—taking ownership of the entire journey from problem definition to solution design to productization.
A platform that digitizes optical shop operations, replacing manual workflows with
a fully integrated system featuring a digital prescription chart and an AI-powered myopia prediction model.
This project was planned, architected, designed, developed, deployed, and operated entirely solo (end-to-end).
- TypeScript-based monorepo (server / front / ai) for unified management
- Flexible Node.js architecture:
- Shifted from 3-layer (Controller–Service–Repository) to
2-layer (Controller–Service) depending on development needs
- Shifted from 3-layer (Controller–Service–Repository) to
- Improved code consistency and overall development productivity
- Stateless authentication using JWT + HttpOnly cookies
- Implemented Two-Factor Authentication (2FA)
- Designed refresh-token management for multi-device concurrent sessions
- Optimized AWS EC2 Free Tier resource usage
- Integrated AI model training and Python ETL into the CI/CD pipeline
- Operated three WAS servers on a single EC2 instance
- Built a MongoDB → Python ETL pipeline
- Developed an XGBoost-based time-series myopia prediction model
- Provided a FastAPI prediction API (JSON) for real-time inference
A system designed to centralize and streamline scattered maritime regulations and
internal rule documents, providing workflow automation and robust document lifecycle management.
- Designed and implemented an electronic approval workflow for regulation changes
- Automated approval processes to reduce manual oversight
- Implemented CKEditor-based web document editing for complex maritime regulations
- Enabled full online creation, editing, and management of regulatory documents
- Added new/old regulation comparison functionality
- Designed file versioning schemas to strengthen change-history tracking
- Improved large-scale document search using MongoDB Atlas + custom search algorithms
- Achieved near real-time search response for thousands of documents
- Migrated the entire legacy JavaScript codebase to TypeScript
- Significantly improved maintainability and runtime stability
- Backend: Node.js(Express), Nest.js, Python(FastAPI)
- Frontend: React(Vite), Nextjs, Zustand, TanStack Query
- Database: MongoDB (Mongoose), PostgreSQL (with TimescaleDB)
- DevOps: AWS (Route53, Load Balancer, EC2, S3, CloudFront, Amplify), Git Actions, Nginx
- AI & Data: Pandas, Scikit-learn, XGBoost
"Building a better world through efficient code."
- Email: dnqkr18@gmail.com