Building portfolio + blog CMS with unified content schema.
Improving editorial workflow and publish guardrails.
About
I work across product, backend, and infrastructure to deliver maintainable services and practical user value.
Skills
Projects
A user-tailored newsletter service, powered by LangChain.
Experience
Full Stack Developer
2026-01 - Present
- Next.js
- Flask Backend Structural Refactoring
Backend Developer
2024-08 - 2025-12
- Developed backend systems for Lightlink, a church management app for Kwanglim Church in Seoul, South Korea.
- Implemented data storage solutions to optimize app performance and user experience.
- Collaborated with cross-functional teams to integrate new features and improve overall functionality.
Links
Tutoring
Build Computational Thinking with Python
In the AI era, what matters most is not memorizing syntax but developing strong logical thinking.
The goal is to help you stop fearing code and know exactly where to start, from search/debug habits to practical automation mini-projects and version-controlled project delivery.
Book a ConsultationLatest from the blog
View all
Getting a response from GenAI is quick and straightforward. But what about the confidence level for that response? In certain applications, especially in the financial domain, confidence scores are required. In a document-parsing task related to financial automation, we tested three approaches to address confidence level: calibrator models, logarithmic probabilities (logprobs), and majority voting. Majority voting turned out to be the most performant technique. Although the idea is relatively simple, the devil is in the details.
When we kicked this off, we weren’t trying to ship an “AI feature.” We were trying to fix a structural problem in how our ads business actually runs in software.