I work on developing real-world systems that use AI/ML approaches. Real-world systems have assumptions that are often different from those of AI/ML approaches. My works aims at 1) designing new system that uses AI/ML approaches; 2) improving the AI/ML approaches leveraging the assumptions in the system. All my work involves several months of implementation followed by thorough testing in real-world data.
Recently, I am striving to redesign genomics software using AI/ML approaches. In the past, I worked on various topics in image super-resolution based on neural network and video delivery system.
Awards and Honors
|Feb, 2022||SAMSUNG PhD Scholarship|
|Spring, 2021||KAIST Breakthrough of the Year 2021, Spring (LiveNAS, NEMO)|
BioinformaticsBWA-MEME: BWA-MEM emulated with a machine learning approachOxford Bioinformatics, 2022SIGCOMMLiveNAS - Neural-Enhanced Live Streaming: Improving Live Video Ingest via Online LearningIn Proceedings of ACM SIGCOMM conference, 2020
CoNEXTCo-optimizing for Flow Completion Time in Radio Access NetworkIn Proceedings of the 18th International Conference on emerging Networking EXperiments and Technologies, 2022SIGCOMMNeuroScaler: Neural Video Enhancement at ScaleIn Proceedings of ACM SIGCOMM conference, 2022EuroSysTowards Timeout-less Transport in Commodity Datacenter NetworksThe 16th European Conference on Computer Systems, 2021MobicomNEMO: Enabling Neural-enhanced Video Streaming on Commodity Mobile DevicesThe 26th Annual International Conference on Mobile Computing and Networking, 2020OSDINeural Adaptive Content-aware Internet Video Delivery13th USENIX Symposium on Operating Systems Design and Implementation, 2018
Adapting AI & ML to SystemSystem Design and Development
https://github.com/kaist-ina/BWA-MEME - Short-read alignment software, released for benefit of research community.
SkillsProgramming Languages: Python, C, C++, SQL, UNIX shell scriptingFrameworks: Tensorflow, Pytorch, Spark, Hadoop, Python Django, Cloudstack