AI Researcher & Engineer
I research and build practical AI systems, with a focus on computer vision. This portfolio includes selected projects and the parts I contributed to.
Work
Projects in applied AI and computer vision.
Applied AI · Semiconductor Inspection
Camera-based inspection for semiconductor manufacturing equipment, covering wafer handling, nozzle tracking, chemical-flow checks, and single-camera pose estimation. I developed and tested computer-vision methods for the individual inspection tasks.
View wafer inspection project →
Applied AI · Webtoon Localization
A modular pipeline for detecting, translating, removing, and reinserting text in webtoon artwork. I worked on the AI research and engineering, including computer vision, OCR, translation, inpainting, and text layout.
View TOONTRA project → View on GitHub ↗Client Delivery Project
Delivered a real-time CCTV monitoring solution for school zones that detected pedestrians entering unsafe road areas and triggered alerts.
Personal Project
RAG-based Korean grammar assistant that searches dictionary entries and finds related examples in past TOPIK reading exams.
Freelance Project
Customer inquiry assistant that answered recurring travel questions using the agency's package and internal information.
Internal Generative AI Tool
Fine-tuned Stable Diffusion 3.5 on a large webtoon dataset and developed an internal one-click LoRA workflow for adapting the model to new characters from a small image set.
I’ve worked on a range of AI projects, with most of my experience in computer vision and generative AI. My work has included image and video analysis, OCR, machine translation, diffusion models, and retrieval-based applications. Computer vision is still my main area of experience, but I enjoy taking on new problems and working in other areas of AI as well.
These case studies describe work I can discuss publicly. Company-owned code, internal datasets, production details, and customer information are not included.