AI & Computer Vision Engineer focused on turning vision pipelines, LLM interfaces, and full-stack product ideas into usable tools.
I build across the stack: OCR-backed web apps, terminal LLM tooling, Discord automation, synthetic data experiments, and Unity/C# game projects.
- Computer vision, OCR, and synthetic image data
- LLM developer tools and agent workflows
- Full-stack web apps for real user workflows
- Python, JavaScript/Node.js, C#, C++, and Unity
| Project | What it shows |
|---|---|
| AjouLLM CLI | Node.js CLI/TUI for model access, project context, planning, and agent-style file edits |
| Once Human Memetic Note | Full-stack game utility with OCR, Discord auth, hive sharing, chat, and trading features |
| Discord TTS | Python TTS helper library for Discord bots with multilingual playback and filtering |
| Doc3D Render | Synthetic document rendering experiments and Blender/Python pipeline fixes |
| Baekjoon Algorithm | C++ algorithm practice and problem-solving notes |
| AppJam Railroad | Award-winning Unity roguelike train game built during the 19th AppJam |
- Portfolio: https://leedonghyun.kr
- GitHub Pages: https://gksmfahd78.github.io

