Industrial Software Leaders Build Secure Autonomous AI Engineers with NVIDIA NemoClaw

Accelerated computing has dramatically reduced simulation time, but challenges remain in end-to-end workflows including CAD, meshing, simulation setup and debugging, and post-processing. NVIDIA and numerous engineering software companies are introducing autonomous AI agents to automate these processes.

These AI engineers are built on NVIDIA NemoClaw, an open blueprint for building specialized, long-running agents equipped with secure runtimes and state-of-the-art models. NemoClaw can integrate with various orchestration frameworks such as OpenClaw and Hermes, and provides the NVIDIA NeMo library and a model router for custom configuration.

Security is handled by NVIDIA OpenShell, an open source runtime. OpenShell controls each agent's access to files, networks, and tools, enforcing policy-based security at every layer. Users can easily deploy NemoClaw via the NVIDIA DGX Spark personal AI supercomputer, enterprise data centers, or cloud services.

Use cases from major industrial software leaders include the following:

• Cadence: Using NemoClaw to build an autonomous RTL engineer, reducing RTL verification time—a core part of digital circuit design—from weeks to hours.
• Dassault Systèmes: Productizing the 3DEXPERIENCE Agentic Platform for design, simulation, and manufacturing operations, built on NemoClaw and OpenShell.
• Siemens: Integrating NemoClaw and OpenShell into its Fuse EDA AI Agent, which plans and orchestrates semiconductor and PCB design workflows.
• Synopsys: Using a NemoClaw-based AI engineer to perform meshing, simulation, and optimization for GPU electronics cooling design with Ansys Icepak.