I am Sungjin Yun, an undergraduate student majoring in Earth and Environmental Sciences, with an academic focus on artificial intelligence, particularly generative models.

My primary research interests include model compression and fine-tuning methods aimed at enhancing the efficiency and adaptability of large-scale generative architectures.

I have developed a fabric texture estimation model using PyTorch and implemented a virtual try-on pipeline utilizing diffusion models and LLm.

Currently, I am building a retrieval-augmented generation (RAG) chatbot using the Ollama framework to deepen my understanding of large language model integration and deployment.

In the long term, I aim to contribute to the development of on-device AI systems and supporting frameworks that are both resource-efficient and context-aware.

I place particular emphasis on designing scalable solutions that remain effective under real-world deployment constraints.