Skip to content
View minseoc03's full-sized avatar
1 follower · 2 following

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report minseoc03

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
minseoc03/README.md

🚀 Minseo Choi | Undergraduate CS @ Johns Hopkins University

I’m an undergraduate CS student at Johns Hopkins University with a strong interest in ML Systems, GPU programming, and LLM inference optimization.

I focus on understanding how ML models run efficiently on modern hardware — from attention kernels and KV cache management to compiler-level and system-level optimizations.
Most of my learning is driven by hands-on implementation, self-study, and technical writing.

🔗 Blog (self-study & technical write-ups): https://minseoc03.github.io
🔗 GitHub: You’re already here 🙂

🔬 Current Interests

  • ML Systems & Inference Infrastructure
  • GPU Programming (CUDA, Triton)
  • Attention Kernels & KV Cache Optimization
  • Compiler & IR-level Optimization (LLVM / MLIR)
  • Multi-GPU & Performance Engineering

🛠️ Tech Stack

Languages

  • Python
  • C / C++
  • CUDA (learning & experimenting)

ML / Systems Frameworks

  • PyTorch
  • Triton
  • LLVM
  • Optuna
  • Hydra
  • scikit-learn
  • NumPy, Pandas, OpenCV, Matplotlib

Tools

  • VS Code
  • Visual Studio
  • Git / GitHub
  • Linux (Ubuntu / Arch-based)

✍️ Learning Philosophy

I value:

  • Learning by implementation
  • Reading papers → reproducing ideas → analyzing performance
  • Writing to clarify understanding
  • Iterating fast based on feedback

📬 Contact

Pinned Loading

  1. mlsys-self-study mlsys-self-study Public

    Hands-on self-study repository on ML systems, including LLVM, MLIR, CUDA & Triton kernels, and efficient LLM attention mechanisms

    C++

  2. medical-enhancement-triton medical-enhancement-triton Public

    A high-performance medical image enhancement pipeline built with Triton GPU kernels.

    Jupyter Notebook

  3. DL_paper_review DL_paper_review Public

    This repository contains PyTorch implementations of deep learning models from research papers. Each model is implemented from scratch following the original paper.

    Python

  4. ko-en-translation ko-en-translation Public

    This repository contains a Transformer-based model for Korean to English translation. The model was trained on conversational data to provide natural translations between Korean and English languages.

    Python

  5. Display-Defect Display-Defect Public

    This repository contains the code and related materials from the LG Aimers competition, where the goal was to develop a machine learning model to predict defective displays in the manufacturing pro…

    Python