Hi, this is Jiawei.
I’m a Physics Ph.D. In grad school, I studied theories and computational methods for quantum physics.
- One topic I worked on is the correlated phenomena emerging from moiré systems. Imagine stacking two layers and giving them a twist — a moiré pattern forms. Because this large-scale pattern shows up in real space, the momentum scale becomes tiny, so small that correlation energies dominate over kinetic energies, and strong correlations appear. I find this system fascinating because it’s an interplay of topology and correlation. To make sense of it, I drew on mathematics to understand the topology, collaborated with my experimental friends, and I get help from CS to actually crunch the numbers. With HPC and different levels of approximation, we try to tame this exponentially complex problem.
- Another topic I worked on is applying machine learning in physics. I’m amazed by the power of AI in daily life, and wondered: can we use it to solve physics problems? One project was to use ML on the vertex function — a Feynman diagram that describes electron interaction. It’s pretty cool that deep learning and data visualization can help uncover the latent structures behind these physical functions. I conducted this research at Columbia University and the Flatiron Institute, where I appreciated the freedom to explore.
I also explored high frequency trading in equities and options at Quantlab and IMC Trading. These experiences sharpened my statistical intuition, teaching me how to extract signals from large, noisy data and giving me a front-row seat to diverse trading styles.
Recently I’ve been studying large language models, very much as a beginner. It gives me a kind of excitement similar to when I first studied physics. This website is a collection of study notes. Here we go.