Email: jialin.utpt@gmail.com

I am a research scientist on the Applied Deep Learning Research team at NVIDIA, working on large language models with a focus on building effective coding agents. Our recent work Nemotron-CORTEXA reached the top of the official SWE-bench leaderboard.

I received my PhD from Caltech advised by Professor Yisong Yue. My PhD research interest was to apply policy learning (reinforcement/imitation learning) techniques to solve optimization problems from combinatorial and Bayesian domains. During my PhD, I interned at Google Brain working on automatic music transcription and reinforcement learning.
I graduated with a bachelor’s degree majoring in computer science and mathematics from University of Toronto where I worked with Professor Michael Brudno, Professor Faith Ellen and Professor Kumar Murty.

For an overview of my PhD research, please see my presentation. [video][slide].

I helped organize the Learning Meets Combinatorial Algorithms Workshop at NeurIPS 2020, which covered a broad range of research in learning to optimize.