AI agents automate chip design, verification, and system development tasks to reduce engineering work, improve designs, and shorten development time.
Synopsys has introduced AgentEngineer, a portfolio of AI agents that can plan and execute engineering tasks across chip design, verification, implementation, and system development. The agents support tasks such as verification coverage closure, software validation, multi-die assembly, and design optimisation.
Built on the Synopsys Autopilot platform, AgentEngineer uses engineering knowledge and design context from Synopsys’ electronic design automation (EDA) and simulation and analysis tools. The platform coordinates tasks across engineering domains and provides controls for security, data access, and workflow execution.
The portfolio includes agents for verification coverage closure, software bring-up and validation, multi-die 3D integrated circuit (3D IC) assembly, and power, performance, and area (PPA) closure. Other agents support analogue layout synthesis, design migration, mask synthesis, and analysis of combustion, resonance, and signal integrity.
The agents use design information, engineering requirements, and workflow data to perform tasks across development stages. They aim to reduce the work involved in engineering tasks, improve design quality, and shorten development cycles.
The Autopilot platform provides tools to coordinate AI agents, manage their skills and memory, monitor their activities, and control engineering workflows. It connects AI systems with design information and engineering tools, allowing them to coordinate tasks and execute actions within defined controls.
The platform supports Synopsys, partner, and third-party infrastructure, AI models, data, agents, workflows, and integrations. Access controls, encryption, and runtime safeguards are designed to protect intellectual property during AI-assisted engineering.
Its context intelligence combines engineering knowledge, tools, reusable skills, and persistent memory to help agents interpret design requirements and carry out tasks. Application programming interfaces (APIs) and design context are also used to reduce AI processing costs and response times.
Synopsys is working with companies to assess how AI agents can automate workflows across design, verification, and system development. The collaborations focus on improving productivity, design quality, and task completion through autonomous engineering.
“Our customers are re-engineering their engineering workflows across semiconductor and systems products to keep pace with increasing system complexity and tight market windows,” said Ravi Subramanian, Chief Product Management Officer at Synopsys.
Click here for the original announcement.
