Siemens Introduces AI Agent for EDA Workflow Automation

[Erkan Teskancan]

## AI Support in EDA Workflows from Siemens

Siemens has introduced Fuse EDA AI Agent, an autonomous system that unifies semiconductor, 3D IC, and PCB design workflows within a single AI-powered framework. This solution manages the entire electronic design automation process to provide automation and efficiency.

### Fuse EDA AI Agent Features

• Workflow-level task planning, coordination, and execution
• Retrieval-augmented generation (RAG) framework providing advanced contextual data access
• Multimodal structure supporting EDA-specific data formats
• Use of Model Context Protocol (MCP) for secure orchestration
• Open integration capabilities with third-party tools

### Comprehensive Semiconductor and PCB Design Support
The Fuse EDA AI Agent supports digital and physical design stages, facilitating front-end development, architectural exploration, RTL design, testbench creation, and debugging processes. As the design progresses, it manages critical steps such as placement and routing, timing closure, and power optimization.

It also provides optimization in power distribution and signal path planning for 3D IC and advanced packaging integration. In PCB development, it supports layout design, signal integrity analysis, and system verification processes.

### Security and Error Reduction in the EDA Process

The Fuse EDA AI Agent is designed to address the industry's complex physical data and IP protection needs. The system offers:

• Built-in verification and guardband mechanisms to reduce design errors,
• Role-based access control and audit trails protecting sensitive data,
• Structured automation with predefined "agent skills" that automate multi-stage engineering tasks.

### Adaptability to Large-Scale and High-Performance Environments

Developed to operate in high-performance computing (HPC) environments, the system optimizes resource utilization and supports hierarchical agent coordination. This allows high-level supervisors to manage sub-agents, ensuring efficient task execution.

Data orchestration is centrally maintained, enabling teams to reuse datasets and perform analyses across workflows. It also maintains compatibility with air-gapped environments common in the semiconductor industry.

### Integration with NVIDIA AI Infrastructure

The Fuse EDA AI Agent works seamlessly with NVIDIA's AI ecosystem, including the NVIDIA Agent Toolkit and Nemotron models. This integration:

• Enhances data access and workflow accuracy,
• Increases the reliability of inter-tool interactions,
• Supports the efficiency of large-scale automation processes.

Thanks to the GPU-accelerated infrastructure, it can process advanced semiconductor and PCB design tasks at high volume and speed.

### A New Era for Autonomous Engineering Workflows in the Industry

The introduction of the Fuse EDA AI Agent signifies a shift from traditional automation to autonomous and domain-focused AI systems. By providing coordinated execution across multiple tools and workflows, it reduces human intervention, shortens design cycles, and leads to better outcomes in semiconductor and PCB system development.