I am a Ph.D. student at the Halıcıoğlu Data Science Institute, UC San Diego, advised by Prof. Jingbo Shang. My research focuses on LLM post-training, continual adaptation, and memory.

I am especially interested in how language models can improve after pretraining: learning from limited task data, rare success signals, long-horizon feedback, and persistent context while retaining prior knowledge. My recent work studies knowledge injection, on-policy self-distillation / reinforcement learning, and memory-based context management.

Beyond research, I am passionate about sports, especially hiking, basketball and motor racing.

I want to express my sincere gratitude to my advisors, collaborators, and mentors for their invaluable guidance and support throughout my research and life.