Jinhwi Kim
  • Home
  • Bio
  • Research & Project
  • Publication & Award
  • Contact
  • Hello, I'm Jinhwi Kim!

    I recieved my B.E. degree from DGIST (Daegu Gyeongbuk Institute of Science and Technology), South Korea in computer science track and in 2025. Also, I'm currently a M.S. student of the same institute in AI major, belonging to DAIL (Distributed AI Lab) under the supervision of Professor Jinhyun So.

    My Research Interests are ...

    Intelligent mobile computing, On-device LLM, Hardware-aware application system design, Reinforcement learning, Federated learning, Edge computing, Dynamic resource management, etc.

    Image
    GitHub Blog YouTube
  • Educaiton

    B.E in DGIST (Computer Science Track)
    M.S student in DGIST (AI, current)

    Experience

    Now, I'm a server manager and technical developer in Godjunpyo AI, a.k.a YoonAI also. I have long-term experiences as a undergarduate student at DAIL (2025.02-2025-08) and ICNL (2020.06-2022.02 and 2024.08-2025.02). Additionally, I taught math and programming (python, C/C++) to many students for past 3 years (2019-22) in a private school or through private teaching, however I'm not keeping the position currently.

    Image Image
  • Image

    On-device LLM System Optimization

    View more
    Image

    Distributed AI Agent Learning

    View more
    Image

    Real-time On-device Translation

    View more
    Image

    Test

    View more
  • Publication

    [1] Jinhwi Kim, Minjae Lee, Seoyong Hong, Janghee Won, Pyeongjun Choi, Jeongho Kwak, Yeongjin Kim, and Jinhyun So, "IGNITE: Thermally-Durable On-device LLMs Governed by Nested Lazy Inference Ignition," submitted for IEEE INFOCOM, 2025.

    [2] Kyungtae Lee, Jinhwi Kim, Jeongho Kwak and Yeongjin Kim, "Dynamic Multi-Resource Optimization for Storage Acceleration in Cloud Storage Systems," IEEE Transactions on Services Computing (TSC), vol. 16, no. 2, pp. 1079--1092, Mar. 2023 (IF: 8.216, JCR Top 1% in Computer Science, Software Engineering). ISSN: 2372-0204.

    [3] Pyeongjun Choi, Jinhwi Kim, Yeongjin Kim, and Jeongho Kwak, "Efficiency Analysis of CPU/GPU Scaling in Embedded AI," in Proc. JCCI, Apr. 2021.

    [4] Jinhwi Kim, Apostolos Galanopoulos, Jude Vivek, and Jeongho Kwak, "A Study on Energy-Process-Delay Tradeoff in Embedded Artificial Intelligence," Proc. ICTC, Jeju, South Korea, Oct. 2020, pp. 1-4.

    Award

    [1] Pyeongjun Choi, Dongho Ham, Jinhwi Kim, Yeongjin Kim, and Jeongho Kwak, "VisionScaling: Dynamic Learning and Resource Scaling in Mobile Computer Vision Applications," Bronze Prize in Samsung HumanTech 2021.

  • Contact

    Please feel free to contact me via email or phone. You can also visit me at DGIST, Daegu, South Korea.

    Email: kjh2159@dgist.ac.kr
    Tel: +82 10 7108 2159
    Office: E3 407, DGIST
    333, Techno jungang-daero, Hyeonpung-Eup, Dalseong-Gun, Daegu, 42988 Korea, Rep. of
Copyright 2025 Jinhwi Kim. All rights reserved.
Designed by Jinhwi Kim