Code0. Portfolio

I’m a third-year undergraduate student at Ajou University, deeply fascinated by computer architecture and AI acceleration. As a research intern in the Embedded and AI Systems Laboratory, I’m exploring how HW/SW co-design can unlock the potential of deep learning systems on specialized hardware.

My journey into this field began with Computer Systems Programming, where I became fascinated with the low-level world of how software commands orchestrate hardware. This experience sparked a deep interest in pushing performance to its limits at the hardware-software boundary. Now, as a research intern, my hands-on work in performance optimization has solidified that passion.

This experience has convinced me that exploring custom accelerator design could be key to achieving better system-level efficiency

Current Research

I’m currently working on HW/SW optimization for Dynamic DNNs, where I’m learning to port and optimize the RISC-V Gemmini systolic array accelerator within CPU-based runtimes like llama.cpp. This challenging project has taught me about the complexities of bridging specialized hardware and software frameworks through deep performance profiling and analysis.

I’ve also initiated the ACE (AI/ML Accelerator Co-design Environment), a personal research endeavor where I’m systematically exploring GEMV (General Matrix-Vector Multiplication) acceleration. Starting from baseline software kernels, I’m progressing through SIMD-optimized CPU backends toward a final hardware implementation of an INT8 systolic array on an FPGA.

Research Interests & Learning Goals

• Hardware/Software Co-design: Understanding how to optimize AI workloads across the full stack
• Computer Architecture: Building a strong foundation through my coursework in Computer Organization, with the goal of exploring RISC ecosystems, systolic arrays, and memory hierarchies.
• AI/ML Acceleration: Learning the fundamental principles of specialized architectures, including key techniques like quantization and structured sparsity.

Technical Skills

• Languages: C/C++ (C++20), Python
• Core Concepts: Computer Architecture, Dynamic DNNs, Low-Level Optimization, Performance Profiling
• Tools & Platforms: VSCode, llama.cpp/ggml, GDB, Git, QEMU, FPGA development (learning)

Future Aspirations

I’m planning to pursue graduate studies in computer architecture, with a focus on specialized computing systems for AI workloads. My goal is to contribute to the development of next-generation architectures that can efficiently handle the growing computational demands of AI applications.

I’m always eager to learn more and discuss research ideas. Feel free to reach out if you’d like to connect!