Last updated December 5th, 2025.

Contact: gussmith@cs.washington.edu

2025-12-05: Our paper, DUET: Agentic Design Understanding via Experimentation and Testing, was accepted to DVCon 2026! The paper demonstrates the effectiveness of encouraging AI agents to build a deeper understanding of hardware designs via simple experiments using a variety of tools (simulation, formal methods, etc.) We completed this work during my time at ChipStack, before the acquisition by Cadence. Arxiv link forthcoming. (for full timeline, see Timeline)

(See my CV for a full record of my work.)

I’m a recently graduated PhD student from the University of Washington’s PLSE Lab, where I worked with Zach Tatlock and Luis Ceze. After my PhD, I worked at the AI-for-EDA startup ChipStack, which was successfully acquired by Cadence. I currently split my time between contracting projects for YosysHQ in open source EDA tooling and advancing research in programming languages, hardware design, and AI with collaborators at UC Santa Barbara and the University of Washington. All of this work is done under my contracting LLC, Southmountain Research.

My current research goal is to build bridges between the new, fast-moving world of LLM-driven AI and the more traditional fields of programming languages (PL) and compilers. I believe there is a goldmine of research at the intersection of these fields. Most obviously, PL techniques can be used to guarantee correctness of AI-generated code (PL for AI correctness) and AI can be used to automate search processes like automated theorem proving (AI for PL). Even more interestingly, though, I believe the beautiful and powerful abstractions developed by PL practitioners can inspire ways to infer or induce structure within massive, inscrutable LLMs.

A representative paper along this vein is our recently accepted work at DVCon 2026:

DUET: Agentic Design Understanding via Experimentation and Testing. To appear at DVCon 2026. Gus Henry Smith, Sandesh Adhikary, Vineet Thumuluri, Vivek Pandit, Kartik Hegde, Hamid Shojaei, Chandra Bhagavatula. (Arxiv link forthcoming.)

In this paper, we demonstrate the effectiveness of encouraging AI agents to build a deeper understanding of hardware designs by conducting experiments with traditional, deterministic tools like simulators, formal engines, and static analysis tools.

Meanwhile, I also pursue non-AI research in programming languages and compilers. In my PhD, my singular research goal was to demonstrate that compilers can and should be automatically generated from formal descriptions of the hardware to which they compile. This is best captured in my dissertation:

Generation of Compiler Backends from Formal Models of Hardware. Dissertation, University of Washington, 2024. Defense talk. Arxiv link.

And by my major PhD projects, namely Lakeroad, a tool which uses vendor-supplied FPGA primitive simulation models to automatically generate hardware compilers targeting those primitives:

FPGA Technology Mapping Using Sketch-Guided Program Synthesis. ASPLOS 2024. ASPLOS conference talk. Arxiv link. Gus Henry Smith, Ben Kushigian, Vishal Canumalla, Andrew Cheung, Steven Lyubomirsky, Sorawee Porncharoenwase, René Just, and Zachary Tatlock. (The Lakeroad paper.)

I am also the author of similar projects such as Churchroad (paper)—a new FPGA synthesis tool which builds on Lakeroad—and Glenside (paper), a pure, functional, low-level, equality-saturation-friendly language for tensor programs.

I have been lucky enough to mentor many great students in my short career. I currently work with Sijie Kong, who is pursuing an MS with Prof. Jonathan Balkind at UCSB. During my PhD at UW, I worked with Vishal Canumalla (now an MS student at Stanford; GitHub, LinkedIn) and Andrew Cheung (now a PhD student at UCSD; GitHub, LinkedIn). Previously at UW, I also worked with Andrew Liu (now at Jane Street). While at Penn State, I worked with a great team of motivated undergrads: Nelson Troncoso Aldas, Christopher Pratt, David de Matheu, and Tom Kawchak.

I’m lucky to have called Pennsylvania home throughout all the previous chapters of my life, having lived there until 2018 when I started a new chapter in Seattle. I grew up in Mountain Top (in the Wyoming Valley, outside of Wilkes-Barre) and have traveled to/lived in all corners of the state. In my free time, I write fiction and poetry, program, surf, and run.