Session

DescriptionLarge Language Models (LLMs) are becoming increasingly important in the semiconductor industry due to their ability to analyze large amount of data, identify somewhat repeatable patterns, and perform efficient optimization of the complex processes. These technologies can significantly improve various aspects of semiconductor technologies, including design, verification, validation, manufacturing efficiency, quality control, and overall innovation by enabling faster design iterations, better predictive maintenance, and more accurate anomaly detection within the manufacturing process.

In this session, we propose to have three complementary technologies that will cover this spectrum. The first presentation is about the expected technological advancements in the domain of Silicon Design and Automation that would be achievable with the usage of generative AI techniques. The second presenter will highlight the potential of LLMs to revolutionize post-silicon test engineering by enabling intelligent, context-aware automation. Key part to their approach is the adoption of a natural language programming paradigm, that would enable the LLM to reason effectively within a specific domain context. Final talk in the session would focus on the view that applications of Generative AI can substantially improve upon human-driven silicon design methodologies, with the help of an underlying agentic framework with open interfaces.