Building at the intersection of AI, statistical learning, and mathematical modeling with experience spanning LLM systems, Applied AI, Data Science and NLP.
Work Experience
Huawei, Hong Kong SAR
Research & Software Intern Dec 2025 – Mar 2026Worked at Huawei Theory Lab on Algorithms and AI Systems, focusing on LLM inference algorithms and optimal cache management using C++ and CUDA.
Core Problem: How can memory and caching be managed in hybrid architectures that integrate attention and Mamba layers to optimize latency and throughput?
IBS Biomedical Mathematics Group, South Korea
Machine Learning Research Intern Dec 2024 – June 2025Developed Deep learning pipelines for classifying sleep disorders and contributing to two research projects.
Core problem: How do we build architectures that exploit the quasi-periodic, multimodal structure of physiological signals and the underlying sleep stage dynamics and circadian rhythms for robust sleep disorder classification when labeled data is scarce?
Users & Informations Lab, KAIST
Undergraduate Researcher, Project: ChatGPT Voices Aug 2024 -- Dec 2024Built a cultural benchmark dataset for low resource language and evaluated LLMs on language and reasoning tasks. Coauthored research paper accepted into AACL-IJCNLP 2025 Workshop.
Core Problem: How does chain of thought and chain of reasoning for LLMs break when dealing with language specific contexts?
Data Strategy Lab, KAIST
Research Intern Apr 2024 – May 2024Studied fairness in AI; reviewed and researched bias mitigation techniques.
Core Problem: What pre training, in training and post training methods can we deploy in order to prevent AI from permeating biases?
Technical Skills
- Languages: Python, C++, C, F#, SQL, JavaScript, Scala
- Libraries & Frameworks: CUDA, PyTorch, NumPy, Pandas, Scikit-learn, SciPy, Django, FastAPI
- Tools: PostgreSQL, Docker, Shell Scripting
- Software Development: Object Oriented Programming, Algorithm Design, Functional Programming, Concurrency, Multithreading, Parallel Computing, Database Management
- Mathematics & Statistics: Stochastic Modeling (Markov Processes, Birth Death Models, Quasi Steady State), Time Series (State Space Models), Statistical Inference, Linear Models, Convex Optimization
Projects
Sparse Attention CUDA Kernel
Efficient sparse attention kernel with Longformer masking and Flash Attention style online softmax, achieving 1.8× speedup and O(1) memory complexity.
Volatility Inference with SDEs & Data Assimilation
Estimated cryptocurrency rolling volatility using a mean-reverting stochastic differential equation with online Bayesian filtering. Results were benchmarked against GARCH(1,1) and GARCH(2,2) models under a strict out-of-sample evaluation setup.
with Data Assimilation
2) Estimate the state online using Kalman and Particle Filter updates.
3) Benchmark final estimates against standard GARCH baselines.
SignalCraft: Crypto Alpha Discovery System
Developed a modular pipeline to identify and backtest predictive signals from crypto spot data. Achieved ~52% hit rate and Sharpe ratio > 1 using ensemble models and adaptive trading strategies.
GitHubYut AI: Korean Traditional Board Game
Built a game AI agent using minimax lookahead with Bayesian optimization, achieving 56.5% win rate with 8 optimized policy weights.
rate
tuned
AI Agent for Automated File Management
Real-time file sorter using semantic embeddings to automatically classify and organize files, achieving 40% efficiency gain and 60% time savings.
gain
saved
Honors & Awards
Leadership and Volunteer Experience
Tutor
Tutored KAIST freshmen in Calculus and Programming; mentored students for math olympiads.
Country Representative
Handled student finances and onboarding; facilitated communication with faculty and student body.
Art of Problem Solving (AoPS)
Olympiad Math Grader 2023 – PresentGraded advanced mathematics olympiad problems and provided detailed feedback to students participating in math competitions.
Mercor
AI Reasoning Expert & Mathematical Problem Designer Sep 2025 – Dec 2025Designed mathematical reasoning problems for IMO Frontier Math project. Created challenging problem sets testing advanced mathematical concepts and logical reasoning for AI model evaluation.
Certifications
Bloomberg Market Concepts (BMC)
Comprehensive financial markets education covering Economics, Currencies, Fixed Income, and Equities.
🏆 View CertificateOther Interests
Competitive Chess
Top 1% on Chess.com; winner of national and regional tournaments.