All work
Work from companies, joint research, side projects and school. The four systems have their own detailed pages, with flow diagrams.
Four systems I built and deployed
KISTI project data layer
As the data engineer on a public R&D platform designed by KISTI researchers, I handled collection, loading and serving, and built the model batches and a FastAPI admin interface.
Led to a follow-on main contractImage-data pipeline for joint research with Hitachi
I rewrote everything from machine-vision camera control and capture to dataset preprocessing as a Python pipeline with a PyQt GUI, and handed the design over to two junior researchers.
Deployed on site in the Netherlands, Sep–Oct 2026Genomic variant annotation and risk report system
Building on an algorithm designed with my team lead, I implemented a variant-calling pipeline, SQL annotation over public references, a FastAPI JSON API and the frontend, and deployed them with Docker and Harbor.
Four-stage SQL annotation over public referencesWJMax game server
I handled the server, database and security, running them on FastAPI (Docker, EC2), RDS MySQL and a Cloudflare proxy and WAF. The published version adds tests.
Up to ~2,000 concurrent players during updatesOther things I built
| Project | What I did | Stack |
|---|---|---|
| 3DOS · vision enhancement on an XR headsetNov 2024 · XR Device Content Makerthon finals | At the finals of the 2024 XR Device Content Makerthon (Pangyo Metaverse Hub), team 3DOS was registered with five members of my undergraduate lab, and I handled development, the presentation and the stage demo. I ported Fourier peplography (fog and smoke removal) and photon counting (low-light recovery), methods published in Prof. Myungjin Cho's research, from the lab's MATLAB code to C++ and OpenCV (one teammate helped with part of the port), built them as an Android native plugin (arm64), and ran them in a Unity app on the Metalens 2 headset on its own camera feed, in front of the audience. The slides are public. |
|
My 2022 undergraduate work in vision and ML (depth estimation, a CNN, Smart Mirror) is on the research page, and the life-insurance disclosure tool is one line under Vision Inside on the CV page.