AnaFlow: AI Agents That Explain Their Analog Circuit Designs
Analog chips connect electronics to the real world—but sizing their circuits is still a slow, trial-and-error craft.
AnaFlow introduces a different approach: a team of specialized, LLM-powered agents that read a circuit’s topology, understand design goals, and iteratively tune device parameters with human-readable reasoning. An adaptive simulation strategy runs only the most informative tests, making the process far more sample-efficient.
In demonstrations on two circuits of different complexity, AnaFlow completed sizing fully automatically—unlike pure Bayesian optimization or reinforcement learning setups. It also learns from its own history to avoid past mistakes and converge faster, turning AI into a transparent design assistant for exploration and verification.
If adopted broadly, this agentic workflow could shorten analog design cycles and build trust by showing its work, not just its answers.
Paper: http://arxiv.org/abs/2511.03697v1
Paper: http://arxiv.org/abs/2511.03697v1
Register: https://www.AiFeta.com
#AI #AnalogDesign #EDA #Semiconductors #LLM #ExplainableAI #ChipDesign
