The Fuse EDA AI Agent aims to coordinate multi‑tool workflows spanning design, verification and manufacturing sign‑off.
The system is built to work with NVIDIA’s Agent Toolkit, Nemotron models and AI infrastructure, and integrates with Siemens’ electronic design automation (EDA) software portfolio. Siemens says the technology will help engineering teams accelerate development cycles while maintaining design quality.
“Fuse EDA AI Agent represents the next evolution of our Fuse EDA AI system, moving from in‑tool AI capabilities to autonomous, end‑to‑end workflow orchestration,” said Amit Gupta, chief AI strategy officer and senior vice president at Siemens EDA. “We are delivering intelligent automation across the complete EDA lifecycle, enabling our customers to dramatically reduce design cycles while maintaining the highest quality standards. Our open architecture allows customers to integrate their own workflows and models, providing the flexibility required for enterprise‑scale AI deployment.”
The new agent builds on Siemens’ existing Fuse EDA AI platform, which includes a retrieval‑augmented generation (RAG) pipeline, an EDA‑specific data lake, specialised file parsers and support for multiple AI models.
According to Siemens, the agent provides automated planning and execution across each stage of chip and PCB development. This includes architectural exploration, design planning and RTL coding using Siemens’ Catapult software during front‑end design. In digital verification, the agent can assist with testbench creation and debugging through the newly released Questa One Agentic Toolkit.
For physical implementation, the system integrates with Aprisa for tasks such as place‑and‑route, timing closure and power optimisation. It also supports custom design and verification through Solido, hardware‑assisted verification with the Veloce platform and physical verification sign‑off via Calibre.
In 3D IC design, the agent manages power and ground optimisation and automates signal path clustering using Innovator3D IC. For PCB development, it assists with layout, signal integrity analysis and related workflows in Xpedition and HyperLynx. It also supports manufacturing readiness through Tessent for design‑for‑test processes and integrates with Calibre for optical proximity correction.
Siemens argues that typical AI systems struggle with EDA tasks due to the specialised, physics‑based nature of design data and the risk of exposing sensitive intellectual property through inadequate access controls or cloud dependencies.
The company says Fuse EDA AI Agent mitigates these risks through domain‑specific expertise, embedded validation and workflow guardrails. The system uses “Agent Skills” to execute complex, multi‑step EDA tasks and allows engineering teams to customise these processes.
A scalable architecture supports dynamic tool discovery, multi‑agent collaboration and autonomous error recovery, enabling integration across Siemens and third‑party toolchains. The platform is designed for high‑performance computing environments, with centralised data orchestration and support for air‑gapped systems.
Security features include role‑based access control, audit trails and optional human checkpoints to prevent the unintended exposure of proprietary design data.