The Silicon Integration Initiative (Si2) AI/ML Schema/Ontology Working Group has publicly released its AI for EDA Ontology Repository. This open-source release provides the semiconductor design community with a formal framework and validated use cases intended to support the development of agentic EDA systems and autonomous workflows. As the industry reaches a critical inflection point, this initiative aims to facilitate the evolution from traditional point tools toward autonomous workflows that require causal understanding and sophisticated workflow orchestration.
Si2 AI for EDA Ontology Repository and Open-Source Access
The repository is available under the Apache 2.0 open-source license via GitHub. To facilitate experimentation and industry progress, the release includes the EDA ontology in TTL/OWL formats, a full documentation package, and ontology validation collateral. To specifically support agent discovery and reasoning, the repository also includes an MCP server. The release is further supported by documented and validated use cases developed by Arizona State University and Drexel University.
Framework for Agentic EDA Innovation
The ontology serves as a reasoning backbone for agentic systems by providing a formal representation of core concepts, such as cells, nets, and timing paths, alongside their associated rules, constraints, and relationships—such as "affects," "depends on," and "constrains." This structure is designed to enable multi-step reasoning across design flows by capturing domain vocabulary, design tradeoffs, and cause-and-effect relationships. According to Si2, this semantic foundation addresses the essential requirements for context awareness and workflow orchestration as the industry shifts toward autonomous systems.
Collaborative Development and Industry Participation
The AI/ML in EDA Schema/Ontology Working Group is a collaborative effort comprising academic and corporate members. Participating organizations include IBM, NXP, Qualcomm, Synopsys, and Siemens EDA, as well as NC State University, Drexel University, and Arizona State University. By providing this common semantic foundation, the working group aims to accelerate innovation in AI-enabled semiconductor design workflows and foster broader industry collaboration. Si2 is currently seeking additional companies and universities to join the collaborative work via their GitHub site or through official contact channels.
Key Takeaways
- The AI for EDA Ontology Repository is released under the Apache 2.0 open-source license and hosted on GitHub.
- The framework includes an MCP server for agent discovery and reasoning, alongside TTL/OWL ontology files.
- The working group consists of members from Siemens EDA, Synopsys, Qualcomm, NXP, IBM, and several universities.
TechInsyte's Take
This release signals a shift toward standardized semantic frameworks to support autonomous, multi-step reasoning in semiconductor design. For infrastructure leaders, the primary interest lies in whether this open-source ontology becomes the industry standard for agentic EDA orchestration. Executives should monitor the adoption rate among the participating EDA vendors to determine if this framework will simplify the integration of AI agents across disparate design tools.
Source: Businesswire
