Agentic AI Automates RTL Waveform Analysis And Debugging (Columbia, Google)

Investigators from Columbia University, Google, and Google DeepMind have introduced an agentic framework aimed at automating the debugging and analysis of register-transfer level waveforms. The researchers argue that traditional manual verification techniques are increasingly inadequate for modern multi-billion-transistor systems-on-chip that underpin artificial intelligence workloads. Their published paper advocates redesigning electronic design automation infrastructure to support autonomous agent systems in semiconductor verification.