I'm Joonhun Lee, an AI research team lead from 🇰🇷
for now. I'm passionate about translating research-stage
ideas into deployable systems that hold up under real-world
constraints, particularly in finance.
Leading AI research across order execution, market and asset class expansion, LLM-mediated financial signaling, and ML-driven asset allocation — translating research-stage agentic patterns into deployable decision systems under real-world constraints.
Built and deployed reinforcement learning systems for order execution and futures trading, designing policies and evaluation pipelines robust to noisy, non-stationary, and partially observable market environments.
Engineered trading infrastructure and quantitative research systems for digital-asset market making — supporting systematic strategy research and decision analysis under realistic execution constraints.
Drove Office of the CEO workstreams across investor relations, business development, operations, and data/product initiatives during a period of rapid growth — bridging product, data, and operations into actionable decisions.
Formulated a generalized-Gaussian framework for TD-error modeling that captures error-distribution tail behavior, improving uncertainty calibration and risk-sensitive policy learning under noisy, non-stationary environments.
Proposed a representation learning framework that aligns stack-wise latent features using Sinkhorn divergence. This approach enables the model to learn invariant representations across source domains, ensuring robust forecasting accuracy even when target domain data is inaccessible or highly non-stationary.
Co-authored a vision representation framework that rethinks masked image modeling through implicit neural representations. Unlike discrete pixel-based methods, MINR achieves strong robustness to various masking strategies and domain shifts while remaining lightweight and parameter-efficient.
Mechanical Watch, Golf & Tennis