This is where I open-source quant research, ship trading tooling, and occasionally break things 🤣

• 🔭 Currently working as an ETF Liquidity Provider (LP) at Mirae Asset Securities, focusing on Korean-listed ETFs

• 💼 Previously worked as a Quant AI Researcher at Qraft Technologies

• 🎓 M.S. in Artificial Intelligence @ Kyung Hee University · KHU AIMS Lab member

• 🌱 Currently learning: market microstructure, ETF market making, execution, and prediction-market mechanics

• 💬 Ask me about: ETF market making · quantitative trading · market microstructure · quantitative ML

• 👨‍💻 Read more about my work at choiinyeol.github.io

• ⚡ Fun fact: I take numpy.random.seed(42) very personally

• 🎓 B.S. in Software Convergence, Kyung Hee University

• 🎓 M.S. in Artificial Intelligence, Kyung Hee University

I'm an ETF Liquidity Provider at Mirae Asset Securities, focusing on Korean-listed ETFs and market making.

Previously, I worked as a Quant AI Researcher at Qraft Technologies, where I focused on the intersection of machine learning and quantitative finance.

I hold a B.S. in Software Convergence and an M.S. in Artificial Intelligence, both from Kyung Hee University. During my graduate studies at KHU AIMS Lab, my research covered empirical asset pricing, portfolio optimization, deep learning, time-series modeling, and prediction-market microstructure.

My current interests sit at the intersection of ETF market making, market microstructure, execution, systematic trading, and quantitative research.

I believe the most interesting quant problems emerge where models meet actual market mechanics — pricing, liquidity, execution, inventory, and risk.

Most of my exploratory work and side projects live as open-source on GitHub. A few highlights:

• korea-deep-factor — Empirical asset pricing with DNN / RNN factor models. Replicates and extends the modern empirical asset pricing literature with deep architectures.

• Portfolio-Optimization-Deep-Learning-With-Candlestick-Image — Mean-variance allocation conditioned on candlestick chart images. Explores whether visual chart features carry exploitable allocation signal beyond standard tabular features.

• prediction-market-analysis — A framework for collecting and analyzing Polymarket and Kalshi market & trade data. Treats event-market flow as a structured alpha signal.

• snusmic-quant-terminal — A terminal-based quant research workbench.

• RuleBase-VS-TimeSeries-Algorithm — Capstone benchmark of rule-based trading vs. deep time-series strategies.

⏩ and many more

During my graduate research at KHU AIMS Lab, I've worked across empirical asset pricing, mean-variance optimization, candlestick-conditioned allocation, and prediction-market microstructure.

My professional focus has since moved closer to the market itself: ETF liquidity provision, market microstructure, execution, hedging, and trading infrastructure.

I believe the most interesting quant problems sit where models meet actual market mechanics — pricing, liquidity, execution, inventory, and risk.

I keep my hands dirty through quantitative research, trading systems, and data-analysis projects on the side.