The use of AI agents in chip design has been limited to small portions of the design flow. Engineers have been proceeding cautiously to understand how agents work, how to ensure they are delivering consistent results, and how different agents work together. The next step is to extend their role beyond just gluing together coding and tooling functions. Mehir Arora, head of engineering at ChipAgents, explains what happens when the role of AI agents is expanded to include semantics and physics, what’s happens when agents work in parallel on different engineering problems, and what’s needed to achieve a 10X or greater improvement in productivity.
https://www.youtube.com/embed/WMKiQUYQhnA
The post AI Agent Orchestration For ASIC Autonomy appeared first on Semiconductor Engineering.
Source: https://semiengineering.com/ai-agent-or ... -autonomy/