NVIDIA Partners With Cadence, Siemens, Synopsys to Build Chip Design AI Agents

NVIDIA

Mar 18, 2026 · Updated Jun 21, 2026

NVIDIA announced at GTC that Cadence, Dassault Systemes, Siemens, and Synopsys are building NVIDIA-powered AI agents to plan, optimize, and verify chip and system design workflows using NeMo, Nemotron models, and CUDA-X.

NVIDIA announced at GTC 2026 that four industrial software leaders are building autonomous AI agents on its platform for chip and system design. Cadence's ChipStack AI SuperAgent handles semiconductor design and verification including testbench coding and debugging. Dassault Systemes is building role-based agents called Virtual Companions on its 3DEXPERIENCE platform. Siemens' Fuse EDA AI Agent orchestrates the semiconductor and PCB workflow from design to manufacturing sign-off, and Synopsys is building AgentEngineer, a multi-agent framework for semiconductor and systems design.

This marks an agentic shift in industrial engineering -- long-running AI agents can now streamline and orchestrate complex design, engineering, and manufacturing workflows. Honda runs aerodynamic simulations 34x faster on the Grace Blackwell platform, and MediaTek accelerates chip design 6x with NVIDIA H100 GPUs.

Evaluate how your chip design or engineering toolchain maps to these agentic platforms -- each vendor now offers a dedicated agent solution spanning the semiconductor lifecycle.

View the full update on nvidianews.nvidia.com

NVIDIA Newsroom

@nvidianewsroomMar 16

Physical AI and autonomous AI agents are reinventing the world of manufacturing. #NVIDIAGTC news: @Cadence, @Dassault3DS, @Siemens, and @Synopsys are building NVIDIA-powered AI agents to plan, optimize, and verify complex chip and system workflows. 🔗 https://t.co/FnErnVtZPe https://t.co/5GACbQJ4BX

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from NVIDIA →

Keep reading

NVIDIA Launches Nemotron Coalition to Build Open Frontier Models With Eight Partners

NVIDIA announced the Nemotron Coalition at GTC 2026, uniting Mistral AI, Cursor, Perplexity, LangChain, and four other labs to build open frontier foundation models. The first base model, codeveloped with Mistral AI, will be open-sourced as the Nemotron 4 foundation.

LangChain Adds NVIDIA Nemotron 3 Ultra for Faster AI Agents

LangChainJun 7

LangChain Adds NVIDIA Nemotron 3 Ultra for Faster AI Agents

LangChain announced immediate support for NVIDIA Nemotron 3 Ultra, an open frontier model designed for long-running AI agents. This integration makes the model's 5x faster inference and up to 30% lower cost for complex agentic tasks directly available to developers using the LangChain framework.