Cheuk-Yiu Chan

I am an MPhil graduate in Electrical and Electronic Engineering from The Hong Kong Polytechnic University, and I love everything related to AI, data, and quant. I am currently building my own end-to-end quant research infrastructure and running live research workflows.

My research focuses on building practical machine learning systems with strong statistical foundations. I have worked on diffusion-based generative modeling and sports trajectory modeling for action prediction, with publications in IEEE TIP, IEEE TCE, and ACM MMSports.

Alongside academic research, I am building end-to-end quantitative research and trading infrastructure. My current project, ForgeBook, combines time-series databases, event-driven backtesting, and machine learning workflows to explore signal extraction and alpha discovery in financial markets.

I enjoy connecting theory with implementation: from probability and optimization to production-ready systems. That includes building LLM-agent harnesses for alpha research, with a focus on tool-calling reliability, verification loops, and reproducible evaluation.

I also have hands-on experience with local LLM deployment and debugging workflows (vLLM, tool calling, and orchestration in practice), and I regularly share practical debugging notes with the community.

I am looking for opportunities to work on challenging problems at the intersection of AI, quant research, and large-scale data engineering.

Please feel free to contact me on LinkedIn or via email if you have any questions about my work.