Summary
Jemin Lee is an assistant professor and applied AI researcher with 12 years of experience bridging academic research and industrial R&D in model compression, compiler-level optimization, and deployment of deep learning models. He has led work at ETRI on quantization, mixed-precision compilers, ONNX- and IR-level optimizations, and co-optimization of pruning and auto-schedulers to produce efficient code for hardware accelerators. His background includes postdoctoral research in context-aware interaction systems and managing large research data platforms, reflecting a rare combination of systems-level ML engineering and human-centered data work. Currently based between San Jose and South Korea, he teaches and mentors graduate students while continuing hands-on research into CNNs, Vision Transformers, and LLM efficiency. Notably, he has experience applying Glow/TVM/Ansor toolchains and DragonBoard-targeted optimization, indicating deep practical knowledge of turning research models into production-ready, hardware-efficient implementations.
11 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Computer Science & Engineering, Doctor of Philosophy (Ph.D.) Computer Science & Engineering at Chungnam National University
Korean, English