Autonomous Electromagnetic Simulation and Modeling of QFN Micro-Electronic Packages Using LLM Agents

This paper presents a novel approach to automating the electromagnetic simulation workflow for packaging using Large Language Model (LLM) agents. We demonstrate the use of LLM-based agents to generate 3D models of Quad Flat No-Lead (QFN) packages in Ansys HFSS based on natural language prompts. The agent autonomously constructs the geometry, sets up the simulation environment, and extracts S-parameters following a full-wave electromagnetic analysis. Subsequently, it interprets the results to generate a compact equivalent circuit model and produces a SPICE-compatible netlist. This end-to-end pipeline enables rapid design exploration, reduced engineering time, and improved accessibility to high-fidelity electromagnetic modeling. The proposed framework lays the foundation for intelligent, prompt-driven design and simulation of systems, bridging the gap between high-level design intent and low-level simulation workflows.