Chunmyong Park (박천명)

I am a Ph.D. Student at Concurrency and Parallelism Laboratory, KAIST School of Computing. I am delighted to be advised by Prof. Jeehoon Kang.

My research interests are in constructing the AI serving systems using heterogeneous processors such as CPU, GPU, NPU, etc. In particular, I am focused on designing resource managers and schedulers for the system to better serve real-world workloads. I am currently working with FuriosaAI for this research. If you want to know more about this research topic, please refer to this link.

Contact

• Email: chunmyong.park@kaist.ac.kr
• GitHub: cmpark0126
• Bibliography: DBLP, Google Scholar
• Place: Rm. 4441, Bldg. E3-1, KAIST (+82-42-350-7878)

Education

• (2021) M.S. in Computer Science. KAIST (supervisor: Prof. Jeehoon Kang).

• (2019) B.S. in Computer Science and Electrical Engineering. Handong Global University.

Publications

Experiences

• Intern, FuriosaAI, July, 2020 - August, 2020.
  • Designing next-generation AI serving system using NPU
• Intern, KakaoBrain, July, 2019 - August, 2019.
  • torchlars: A LARS(Layer-wise Adaptive Rate Scaling) implementation in PyTorch, [github], [blog post]
• Intern, DeepBio, January, 2018 - February, 2018.
  • CAMELYON17 project: Designing the deep learning model for detection and classification of breast cancer metastases, [github]
• Research Assistant, Deep Learning Lab (director: Prof. Injung Kim), Handong Global University, March, 2017 – June, 2019.
  • WICWIU: A deep learning framework in C/C++, [github], [thesis]

Awards and Honors

• Best Paper Award, Korea Computer Congress (KCC), 2018.

Calendar