Cadence Design Systems announced two AI partnerships at its CadenceLIVE event on April 15, pairing Google’s Gemini models with its ChipStack AI Super Agent platform and expanding its collaboration with NVIDIA around physics-based simulation, digital twins, and robotics, according to Artiverse and Forbes.
The Gemini Integration
Cadence is integrating its electronic design automation tools with Google’s Gemini models on Google Cloud to automate chip design and verification workflows. The ChipStack AI Super Agent uses model-based reasoning with native design tools to coordinate tasks across multiple chip design stages, interpreting design requirements and executing them without sequential human handoffs.
A new agent targets physical layout processes, translating circuit designs into silicon implementations. This builds on an earlier agent introduced this year for front-end chip design, where circuits are defined in code-level descriptions. Together, they cover the full design pipeline from logic to layout.
The cloud deployment lets semiconductor teams run design workloads without on-premise compute infrastructure. Cadence reported productivity gains of up to 10x in early deployments across design and verification tasks, though the company did not disclose specific customer implementations, according to Artiverse.
“We help build AI systems, and then those AI systems can help improve the design process,” Cadence CEO Anirudh Devgan said at the event.
The NVIDIA Expansion
The NVIDIA partnership focuses on integrating Cadence’s multi-physics simulation and system design tools with NVIDIA’s CUDA-X libraries, AI models, and Omniverse simulation environment. The tools model thermal and mechanical interactions so engineers can assess how systems behave under real-world operating conditions before physical deployment.
The collaboration extends beyond chip design to cover networking, power systems, and robotics development. Cadence’s physics engines are being linked with NVIDIA’s AI models for training robotic systems in simulated environments, reducing the need for real-world data collection.
“We’re working with you in the board on robotic systems,” NVIDIA CEO Jensen Huang said during the event, as reported by Artiverse.
Industrial robotics companies including ABB Robotics, FANUC, YASKAWA, and KUKA are already integrating NVIDIA’s Isaac simulation frameworks and Omniverse-based digital twin tools into virtual commissioning workflows.
Why This Matters for the Semiconductor Stack
Chip design has historically been one of the slowest, most expensive layers of the technology stack, with design cycles measured in months and verification consuming more engineering hours than design itself. Cadence is now applying the same agentic automation pattern that software companies use for code generation and deployment to semiconductor engineering.
The competitive dynamic is straightforward: teams using agent-driven design that compresses cycles and improves verification coverage will ship silicon faster than those relying on traditional sequential workflows. With Altera, NVIDIA, Qualcomm, and Tenstorrent among the early deployment partners, according to Business Wire, the question for the rest of the industry is whether they can afford to wait.