Siemens Digital Industries Software has expanded its EDA collaboration with NVIDIA to add agentic AI capabilities for electrothermal analysis in 3D IC design, targeting one of advanced packaging’s more complex design challenges. The update combines Siemens’ self-verifying agentic AI technology introduced at DAC 2026 with NVIDIA’s accelerated computing platform. The workflow integrates electrical, power and thermal analysis into a single automated process for evaluating design tradeoffs before signoff. Rather than coordinating multiple tools manually, engineers define design objectives while the AI agent manages coupled electrothermal analyses, identifies failed targets and traces issues to factors such as power delivery, die placement or package materials. The system then explores alternative configurations and reruns promising candidates against design goals. Siemens said the workflow uses its Design Space Exploration and Optimization technology to evaluate hundreds of design variants with physics-based signoff solvers. In benchmark tests, it reduced engineering effort by up to 95%, cut workflow setup time by up to 99% and increased design exploration coverage by as much as 50 times compared with conventional manual flows. “Electrothermal codesign has traditionally required experts to coordinate multiple tools and manually iterate through a limited set of design alternatives,” said Ankur Gupta, executive vice president, EDA IC Portfolio, Siemens. “By combining agentic AI with Siemens’ proven physics-based analysis, engineers can now evaluate up to 50 times more design scenarios, enabling faster and more confident decisions for the next generation of AI infrastructure.” The workflow automates nine stages across seven Siemens EDA tools, including Innovator3D IC Integrator, Calibre 3DThermal, Simcenter Flotherm, Calibre xACT, Solido Characterizer, Solido SPICE and mPower Digital. Siemens said a typical run executes 50 to 100 reasoning steps over eight to 10 hours while sweeping 200 to more than 500 design points across power budgets, material stack-ups and heat-transfer conditions. NVIDIA’s contribution includes Agent Toolkit, Nemotron-3.5-Lightning models and CUDA-X libraries, which accelerate supported simulation workloads and shorten iteration time for agent-driven design exploration. “Power delivery and heat dissipation increasingly determine what’s possible in 3D chip design,” said Tim Costa, vice president and general manager of industrial and computational engineering at NVIDIA. “With NVIDIA Agent Toolkit and Nemotron-3.5-Lightning, Siemens’ agents coordinate electrical and thermal solvers to deliver up to 50x greater coverage in design space exploration with 95% less engineering effort. By diagnosing how power delivery, die placement and package materials contribute to missed targets, these agents make electrothermal analysis an integral part of chip development.” The announcement comes as 3D IC packaging becomes increasingly important for AI accelerators, where higher compute density and stacked memory increase power density and concentrate heat, making thermal and electrical co-design an earlier-stage design requirement. The post Siemens extends NVIDIA AI collaboration for automated 3D IC electrothermal design workflows appeared first on EE Times Asia .