I am a Quantitative Researcher at Tower Research Capital (Latour team) in New York, where I design and apply machine learning methods to build predictive signals (alphas) for global financial markets. Prior to joining Latour, I was a Quant at G-Research, where I focused on developing neural architectures and training methods to extract orthogonal signals from large, noisy datasets.
My research bridges AI for financial prediction and fundamental machine learning. It combines two interests: applying AI methods in markets and contributing to core ML research. I have worked on neural network compression, reinforcement learning, and scalable training methods, with a particular focus on connecting structured random embeddings to novel neural representations that trade high dimensionality for convexity.
I hold a PhD in Electrical Engineering from Stanford University, advised by
Mert Pilanci, and have collaborated widely across AI research—developing reinforcement and imitation learning algorithms for safety-critical systems with Marco Pavone, visiting UC Berkeley with Laurent El Ghaoui, interning at Facebook AI Research with Mohammad Ghavamzadeh, and contributing as reviewer for NeurIPS, ICML, ICLR, and other leading conferences.
Outside of research, I’m passionate about mathematics and discovering the best food wherever I go. I’m also an avid runner.