How Does Synopsys Use AI in Chip Verification?
The new agentic workflow also introduces Synopsys' first fully autonomous computer-aided engineering capability for electronics thermal simulation, built with NVIDIA technology.
On July 26, 2026, Synopsys, the electronic design automation (EDA) company, announced new agentic AI capabilities for chip and electronics system design at the 2026 DAC Chips to Systems Conference. The company said the advancements were developed in collaboration with NVIDIA and are enabled by NVIDIA Nemotron, running on the company’s accelerated computing platform and secured by the NVIDIA OpenShell runtime.
"AI is fundamentally reshaping engineering, and Synopsys is at the forefront of this transformation, enabling fully autonomous agents across every stage of silicon and systems development," said Ravi Subramanian, Chief Product Management Officer at Synopsys.
According to the company, the centrepiece of the announcement is a fully autonomous, long-running design verification (DV) agent that orchestrates the entire chip verification cycle. The company said the workflow shifts verification from a tool-assisted process to a goal-driven one, with an orchestrator agent deconstructing verification goals from specification, design, test repository, and user inputs
The orchestrator then coordinates specialized agents across test plan generation, coverage closure, and debug, the company said.
Synopsys said the technology is built on its AgentEngineer platform combined with NVIDIA's Agent Toolkit, the Nemotron 3 Ultra open model, and the OpenShell runtime.
Enhanced Features
Synopsys reported that the autonomous DV workflow delivers up to 50 times faster time-to-validated RTL (register-transfer level) design, along with a 20% improvement in coverage, compared to traditional verification workflows.
The company also cited a separate agentic workflow for analog and mixed-signal design, powered by its Custom Compiler Layout Synthesis tool, which it said improves productivity by up to 3 times.
For the thermal management workflow, built with Ansys Icepak electronics cooling simulation software (now part of the Synopsys portfolio), Synopsys said the agentic system autonomously handles simulation setup, pre-processing, and post-processing in a fraction of the time required by manual methods.
Separately, the company said its PrimeSim SPICE circuit simulator now runs up to 18 times faster using NVIDIA GPUs, part of a broader portfolio of more than 20 GPU-accelerated Synopsys products spanning physical verification to photonics simulation.
Synopsys said customers are currently evaluating the new agentic capabilities, with general availability planned for the second half of 2026. The company is demonstrating the workflows this week at its booth during the DAC Chips to Systems Conference.
Tim Costa, Vice President and General Manager for Computational Engineering at NVIDIA, said the collaboration reflects a broader shift in engineering. "The future of engineering is agentic, where AI agents reason, plan, execute complex workflows and verify their own work across the entire product development lifecycle," Costa said.
>
>
Key Takeaways
- Synopsys launches autonomous AI agents for chip verification, enhancing electronic design automation.
- Leverage NVIDIA technology to achieve faster, goal-driven verification processes in chip design.
- The new workflow can validate RTL designs up to 50 times faster and improves coverage by 20%.
- Synopsys' advancements are based on its AgentEngineer platform and NVIDIA's Agent Toolkit.
- AI is transforming engineering practices, positioning Synopsys as a leader in the EDA sector.