Hi! Welcome to Hanxian Huang’s webpage

I am a software engineer at Meta, working on efficient LLM deployment.

Before joining Meta, I received my Ph.D at the Computer Science and Engineering Department of UC San Diego, where I was fortunate to be advised by Prof. Jishen Zhao. Before joining UCSD, I received my B.S. degree from the School of Electronics Engineering & Computer Science at Peking University, advised by Prof. Guojie Luo. My current research broadly spans and stretches the intersection of machine learning (ML) with programming languages, compilers, and computer systems. This includes exploring advanced ML techniques to improve system designs and PL tasks, as well as efficient ML algorithms and system co-design.

News

2025/02: Our paper, “MAGE: A Multi-Agent Engine for Automated RTL Code Generation”, is accepted to the Design Automation Conference (DAC) 2025! Try MAGE here.

2024/12: I have passed my Ph.D. final defense and earned my Ph.D. from UCSD! I will be joining Meta to work on efficient LLM deployment! Looking forward to this exciting new journey!

2024/10: Our summary of “AI for Chip Design” tutorial at Hot Chips 2024 has been published on the ACM SIGARCH technical blog. Thanks for all the collaborators for putting them together!

2024/09: Our ISSTA’24 paper “Multi-modal Learning for WebAssembly Reverse Engineering” was selected as one of 11 (out of 143) papers to receive an ACM SIGSOFT Distinguished Paper Award!

2024/08: I will give a talk on “Domain Adaptive LLM Models for Chip Design” at the Hot Chips 2024 tutorial, see you at Hot Chips!

2024/08: I was invited as a speaker at IEEE MIPR Innovation Forum Industry Challenges of Efficient AI.

2024/07: I was invited to the ML and System Rising Stars workshop [poster]! Thanks ML Commons for this great organization and support!

2024/06: Our workload forecasting project was highlighted in the Microsoft Research Blog!

2024/05: I was selected as one of the Machine Learning and Systems Rising Stars 2024 by ML Commons! See you at the workshop in the NVIDIA Headquarters on July 15-16!

2024/05: Our paper, “Learning to Maximize Mutual Information for Chain-of-Thought Distillation”, is accepted to the ACL Findings 2024!

2024/04: Our paper, “Fasor: A Fast Tensor Program Optimization Framework for Efficient DNN Deployment”, is accepted by the International Conference on Supercomputing (ICS) 2024!

2024/03: I’ll be working as a research scientist intern in the Pytorch team at Meta this summer and looking forward to meeting you in Menlo Park!

2024/03: Our paper, “Multi-modal Learning for WebAssembly Reverse Engineering”, is accepted by The ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) 2024!

2023/11: Our paper, “Sibyl: Forecasting Time-Evolving Query Workloads”, is accepted by the ACM SIGMOD Conference 2024!