Kyu Song
Applied Scientist, Member Of Technical Staff at Mirage
New York City Metropolitan Area United States
Summary
👤
Senior🎓
Top SchoolKyu Song is an applied scientist with nine years of experience building and scaling multimodal AI systems, currently focused on high-throughput video generation at Mirage. He has led teams and research programs—from heading an 18-person AI group at MINDsLab to driving edge foundation model research and event detection with VLMs showcased at NVIDIA GTC. Kyu’s work spans speech synthesis, real-time multilingual talking-face generation (25 fps on consumer GPUs with a popular Hugging Face demo), and efficient diffusion models using distillation and pruning presented at ICCV and ECCV. He combines hands-on model engineering with product delivery, contributing to a company public listing and practical multimodal ID systems early in his career. Based in the NYC area with a KAIST background in Bio & Brain Engineering and CS, he also brings cross-disciplinary interests in synthetic biology and optogenetics hinted at in his GitHub bio. Known for marrying research rigor with production constraints, he excels at squeezing state-of-the-art models into real-world, resource-constrained deployments.
9 years of coding experience
6 years of employment as a software developer
Undergraduate, Undergraduate at Korea Advanced Institute of Science and Technology (KAIST)
Korean, English
Stackoverflow
1reputation
0reached
0answers
0questions
Programming languages (1)
Github contributions (5)
Kyuye/kyuye.github.io
Mar 2020 - Sep 2023
Contributions:198 pushes, 1 branch in 3 years 6 months
MLJejuCamp2017/REGAN_Kyuye_Rachel
Jul 2017 - Jul 2017
Contributions:49 commits, 43 pushes, 1 branch in 22 days
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.