Abstract
The design of modern semiconductors face challenges with numerous parallel place and route runs triggered by continuous updates, which are utilized to achieve quality of results and power, performance, and area convergence. The described technology introduces an artificial intelligence-assisted physical design convergence cockpit, which operates as a multi-agent system powered by large language models and retrieval-augmented generation engines. This system acts as an autonomous orchestrator that monitors design repositories, intelligently launches and tracks electronic design automation flows, performs real-time failure triage by parsing log files, and executes self-healing design closures by generating corrective scripts. The system features a modular architecture that effectively decouples the core execution environment from its specialized intelligence layers, ensuring the system can adapt to new capabilities without disrupting workflows. Transitioning from manual tool operation to autonomous task orchestration reduces engineering effort and accelerates development schedules.
Keywords: artificial intelligence, physical design, convergence, multi-agent system, large language models, retrieval-augmented generation, electronic design automation, self-healing, autonomous orchestrator, modular architecture, deterministic data extraction.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
N/A, "Artificial Intelligence-Assisted Physical Design Convergence Cockpit", Technical Disclosure Commons, (April 01, 2026)
https://www.tdcommons.org/dpubs_series/9694