Computer Systems and AI Lab (CoSAIL)
Computer Systems and AI Lab (CoSAIL)
Embedded Files
컴퓨터 시스템 및 AI 연구실 석사/박사 과정 및 학부 인턴에 대해 궁금한 학생들은 mhan@knu.ac.kr 메일으로 연락 주세요.
CoSAIL@KNU
CoSAIL@KNU
Welcome to the Computer Systems and AI Lab (CoSAIL) in the School of Computer Science and Engineering at Kyungpook National University. Our research lies at the intersection of computer systems and AI, with a broad focus on system-level challenges.
Specifically, we explore:
Systems for AI – Designing computer systems for efficient AI computing
AI for Systems – Leveraging AI techniques to optimize computing systems
System Software for Emerging Hardware – Building system SW for next-generation hardware (e.g., NPUs, Processing-in-Memory)
The following keywords describe our research interests: Operating Systems, Computer Architecture, AI Systems, AI-augmented Computer Systems, and Data Center / Cloud Computing. We are actively looking for graduate and undergraduate students to join our team. If you are interested in our research group, feel free to contact us (mhan@knu.ac.kr).
Our current research projects include (but are not limited to):
Resource Management Techniques for Efficient On-Device LLM Inference
System Software for On-Device Retrieval-Augmented Generation (RAG)
[On-Device LLM Inference]
[On-Device LLM Inference]
1. Resource Management Techniques
for Efficient On-Device LLM Inference
1. Resource Management Techniques
for Efficient On-Device LLM Inference
Adapting emerging AI techniques (e.g., speculative decoding, MoE) for efficient on-device execution
System-level support for memory management, scheduling, and runtime optimization
Cross-layer optimization of compute, memory, and storage under resource constraints
[On-Device Retrieval-Augmented Generation]
[On-Device Retrieval-Augmented Generation]
2. System Software for On-Device RAG
2. System Software for On-Device RAG
Profiling and characterizing RAG workloads on constrained devices to guide system-level optimizations
End-to-end optimization of coordination between LLMs and vector databases
System-level techniques for memory management, parallelism, latency reduction, and caching
News
News
[Aug. 2025] Prof. Han has been awarded the Outstanding Young Scientist Grants (우수신진연구(씨앗연구)) by the National Research Foundation of Korea (NRF). Our project, “Input Complexity-Aware Adaptive Execution and Memory Management for Efficient On-Device LLM Inference”, will be funded from September 2025 to August 2026.
[Jun. 2025] We’re excited to share that our lab has been selected for the Google Research Credits Program (Jun 2025 – Jun 2026).
[Sep. 2024] The Computer Systems and AI Lab (CoSAIL) at KNU was launched.
[Sep. 2024] Myeonggyun Han was appointed as an assistant professor in the School of Computer Science and Engineering at Kyungpook National University.
Page updated
Google Sites
Report abuse