Andrew B. Kahng

CSE and ECE, UC San Diego

Agentic AI for EDA: Observations and Outlook

Abstract

The landscape of agentic AI for EDA R&D spans (i) tool-integrated agents for
algorithm evolution, debugging, and workflow automation; (ii) autonomous exploration of
heuristic spaces in classical design task domains; (iii) interfaces between agents and traditional
EDA frameworks; and more. Particularly in open source, a first-generation “agentic EDA R&D
software engineer” is rapidly taking shape, via (i) repo-scale, closed-loop autonomous coding
pipelines; (ii) GitHub issue creation and code implementation that can draw on research corpora;
and (iii) bespoke EDA tools and flows that address design-specific optimization targets.
This talk will start with observations on agentic AI for EDA. First, mostly-technical near-term
foci: value propositions and ROI; validations and benchmarking; essential rebalances (humans
vs. agents; search and optimization vs. evaluation). Second, “speed limits”: fundamental time
constants and scaling phenomena that the future may or may not overcome. Third, societal
considerations: IP and creators’ rights; implications of sovereign efforts; and mores of the
(technical) community. Such observations inform an outlook: “what comes next?”.
[Related talks and papers on open-source EDA and AI/ML for EDA can be found at
https://vlsicad.ucsd.edu]

Biography

Andrew B. Kahng is Distinguished Professor of CSE and ECE and holder of the
endowed chair in high-performance computing at UC San Diego. He was visiting scientist at
Cadence (1995-97) and founder/CTO at Blaze DFM (2004-06). He is coauthor of 3 books and
over 500 journal and conference papers, holds 36 issued U.S. patents, and is a fellow of ACM
and IEEE. He was the 2019 Ho-Am Prize laureate in Engineering. He has served as general chair
of conferences such as DAC, ISPD, SLIP and MLCAD, and from 2000-2016 served as
international chair/co-chair of the International Technology Roadmap for Semiconductors
(ITRS) Design and System Drivers working groups. He was the principal investigator of the U.S.
DARPA “OpenROAD” project (https://theopenroadproject.org/) from June 2018 to December
2023, and until August 2023 served as principal investigator and director of “TILOS”
(https://tilos.ai/), a U.S. NSF AI Research Institute.