Math @ KAIST · IMO Bronze '20 '21 · AI @Huawei . Computational Mathematics @IBS
I am interested at the intersection of mathematical modeling, systems, and AI , with a bias toward problems where theory actually constrains what you can build.
- Huawei Theory Lab — LLM inference algorithms, CUDA kernels, KV cache management for hybrid attention/Mamba architectures
- KAIST — 3rd year, Department of Mathematical Sciences , School of Computing
- WorldQuant Brain — Consultant; Alpha research for a long short neutral hedge fund.
| **IBS Center for Computational Matheamtics ** | Deep learning for sleep disorder classification from multimodal physiological time series. Inductive bias design under limited data. |
| KAIST Users & Info Lab | Built cultural NLP benchmark for low-resource languages; evaluated chain-of-thought breakdown across language contexts. Paper accepted at AACL-IJCNLP 2025 Workshop. |
| KAIST Data Strategy Lab | Pre/in/post-training bias mitigation in foundation models. |
Sparse Attention CUDA Kernel
Longformer-style sparse attention with Flash Attention online softmax. 1.8× speedup, O(n·w) complexity, O(1) memory.
Volatility Inference with SDEs
Heston-lite and OU SDEs with Kalman/Particle filters for crypto volatility estimation. R² ≈ 0.9 across both pipelines vs ~0.13–0.48 for GARCH baselines.
SignalCraft
Modular crypto alpha pipeline. ~52% hit rate, Sharpe > 1 with ensemble models and adaptive execution.
Yut AI
Minimax agent with Bayesian-optimized policy weights for Korean board game Yutnori. 56.5% win rate over baseline.
Python C++ CUDA PyTorch NumPy SQL JavaScript F# Scala
PostgreSQL Docker FastAPI Django