Practical computing, acceleration, and real-world system design
This page is used for course materials, code sharing, and small-scale experiments. If you arrived here unintentionally, this is likely a temporary endpoint used during lectures.
The work here focuses on GPU acceleration, efficient model execution, and practical system-level optimization. Background includes embedded systems and low-level performance-oriented design.
Recent exploration includes lightweight adaptation techniques such as LoRA and QLoRA, with an emphasis on making large-scale models more efficient and deployable under constrained environments.
This is not an official lab homepage. It is a minimal interface intended for teaching and quick access during sessions.