Hyunsu Kim
Software engineer, Constant learner
Currently
I am a software enginner at Furiosa AI in Seoul, Korea.
Research Interests
Compiler infrastructure for effective and efficient HW acceleration (MLIR)
Neural network pruning and quantization
Program representation for robust learning
Education
2020 - 2022
Masters student, KAIST
- Programming Systems Lab (advisor: Kihong Heo), Graduate School of Information Security
2014 - 2020
B.S., KAIST
- School of Computing, Mathematical Science (Double major)
2012 - 2013
Gangwon Sci. High School
Publications
ICSE 2022
Learning Probabilistic Models for Static Analysis Alarms
- Hyunsu Kim, Mukund Raghothaman, and Kihong Heo
- π Best Artifact Award
- Proceedings of the 44th ACM/IEEE International Conference on Software Engineering, ICSE 2022, May 2022.
- [paper] [full-version] [video] [artifact]
KCC 2021
λ² μ΄μ§μ μλ λνΉ μμ€ν
μ κ±°μ§ μΌλ°ν λ¬Έμ μ ν΄κ²° λ°©μ
- Hyunsu Kim and Kihong Heo
- π μ°μλ Όλ¬Έμ
- Korea Computer Congress, KCC 2021, June 2021.
DEEM 2019
Data Cleaning for Accurate, Fair, and Robust Models: A Big Data - AI Integration Approach
- Kihyun Tae, Yuji Roh, Young Hun Oh, Hyunsu Kim, and Steven Euijong Whang
- Proceedings of the 3rd International Workshop on Data Management for End-to-End Machine Learning, DEEM at ACM SIGMOD 2019, June 2019.
Talks
2022
Learning Probabilistic Models for Static Analysis Alarms, KCC 2022 Top Conference μΈμ
, Jun 2022.
When Usability met 2FA, BSides Munich 2022, May 2022.
- Hyunsu Kim, Junoh Lee, Myeong Geun Shin, Kihong Heo and Sang Kil Cha.
BayeSmith: μ μ λΆμ μλμ μν νλ₯ λͺ¨λΈ νμ΅, SIGPL Winter School 2022, Feb 2022.
Awards
2022
Best Artifact Awards, ICSE 2022.
2021
μ°μλ
Όλ¬Έμ, KCC 2021, Korean Institute of Information Scientists and Engineers (νκ΅μ 보과νν).
2018
Qualcomm Innovation Awards, Qualcomm-KAIST Innovation Awards Committee.
Occupation
now
Furiosa AI
- Software engineer
2019
Classum
- Web front-end developer
Projects
Check out my github repo: 
.
Last updated: Aug 2022