Summary
Hyeji Kim is a Principal Researcher at ETRI with nine years of experience specializing in AI processors, neural network compression, and low-power embedded SoC design. She blends deep academic training (Ph.D. in Electrical Engineering from KAIST) with hands-on VLSI and FPGA implementation experience, having led projects from tensor-decomposition optimization for CNNs to ASIC stereo depth extraction. Her work spans system-level AI SoC platforms, efficient neural-network compression algorithms, and hardware-accelerated surveillance and communication modules, reflecting a rare cross-domain fluency in algorithms and silicon. Notably, she has delivered energy-efficient turbo decoders and CRC-based early termination units, demonstrating attention to practical low-energy constraints in real deployments. Based in Daejeon, South Korea, she drives research that moves models from algorithmic compression to production-grade hardware. Colleagues describe her as a pragmatic researcher who closes the loop between novel compression methods and their digital VLSI realizations.
9 years of coding experience
6 years of employment as a software developer
Master's degree, Electronic Engineering, Master's degree, Electronic Engineering at Chungnam National University
Ph.D, Electrical Engineering, Ph.D, Electrical Engineering at Korea Advanced Institute of Science and Technology
english, korean