Synopsys and Ansys power the future of innovation—connecting silicon to systems.
EDA
System
IP
SolvNetPlus
Training & Education
For the past several years, AI has helped engineers work faster. Copilots have answered questions, generated scripts, analyzed results, and suggested fixes, while people continued to direct the workflow and make the decisions. Agentic AI expands that role from assisting with individual tasks to executing complete workflows.
Today, we are introducing a portfolio of Synopsys AgentEngineer™ solutions for chip and systems design, built on a trusted platform to power the age of autonomous engineering.
As long-horizon super agents, AgentEngineer solutions can reason through complex engineering objectives, develop and adapt plans, orchestrate specialized task agents, and execute complete workflows across verification, system validation, implementation, analog and mixed-signal (AMS) design, manufacturing, and simulation and analysis. They are built on the new Synopsys Autopilot™ Platform, an open and secure foundation that provides context intelligence, maintains persistent memory, connects agents to engineering tools and data, and enables telemetry and governance for autonomous engineering workflows.
The long-horizon difference
Synopsys AgentEngineer solutions are designed for long-horizon engineering. Unlike long-running agents — which may perform one activity for hours or days, such as monitoring nightly regressions — long-horizon agents address goal complexity. A Synopsys AgentEngineer can pursue an objective that requires hundreds or thousands of reasoning steps while maintaining context, evaluating results, and adapting its plan along the way.
Consider an end-to-end verification flow with a defined specification and coverage goal. Reaching verified RTL can require interpreting the specification, generating RTL, developing the verification plan and testbenches, running tests, measuring functional and code coverage, analyzing failures, correcting the RTL or tests, rerunning jobs, and repeating the process until the target is met.
A Synopsys Verification AgentEngineer can reason through that objective, develop a plan, and orchestrate specialized task agents, including long-running agents, needed at each stage. One task agent might generate a testbench, another might analyze coverage gaps, and another might perform root-cause analysis. The Verification AgentEngineer evaluates their outputs, decides what comes next, and adjusts the workflow when an intermediate result falls short.
A portfolio of specialized AgentEngineer solutions
As part of the broader Synopsys AI solutions portfolio, long-horizon AgentEngineer solutions address complex objectives across chip and systems design:
- Verification AgentEngineer orchestrates the end-to-end design verification workflow, from specification interpretation through coverage closure.
- Implementation AgentEngineer coordinates implementation and signoff across multiple stages, including floorplanning, placement, routing, congestion analysis, design for test (DFT) optimization, and timing, power, and design-rule closure.
- AMS AgentEngineer orchestrates analog and mixed-signal design workflows spanning design and optimization, layout synthesis, IP node migration, physical verification, design closure and signoff, and transistor-level timing and characterization.
- Manufacturing AgentEngineer orchestrates the design-to-silicon manufacturing workflow across process and device simulation, mask synthesis, and mask data preparation.
- Meshing AgentEngineer generates, validates, repairs, and optimizes meshes from user intent using workflow automation, AI reasoning, and built-in meshing best practices.
- Blaze AgentEngineer automates gas turbine combustion studies, determines combustion modeling strategies and flow-regime requirements, and orchestrates end-to-end simulation workflows.
- EMC AgentEngineer automates PCB EMI/EMC analysis, including EMI scanning, electromagnetic extraction, assessment of radiated emissions against applicable EMC limits, and design iteration to help meet EMC compliance requirements.
General availability for Synopsys AgentEngineer solutions and the Autopilot Platform is planned for the end of 2026.
The trusted foundation for autonomous engineering
Advanced chip and systems design cannot be reduced to a sequence of isolated prompts. Long-horizon AgentEngineer solutions need a trusted foundation that gives them the context and tools required to act on their decisions. They also need enterprise controls governing how data, models, agents, and infrastructure are used.
The Synopsys Autopilot Platform provides that foundation. Its shared services supply context intelligence, reusable skills, persistent memory, telemetry, security, and governance while connecting agents to Synopsys engineering knowledge, tools, and data. By combining domain engineering knowledge with reusable skills and memory, context intelligence helps agents guide workflows with greater accuracy while improving token efficiency and reducing latency.
The platform brings together three complementary layers to support autonomous execution:
- Long-horizon AgentEngineer solutions are domain-specific super agents that orchestrate task agents, evaluate intermediate results, and determine what should happen next in pursuit of a complex engineering objective.
- Task agents complete specific, bounded engineering tasks with defined inputs and outputs. They may be orchestrated by a long-horizon AgentEngineer or invoked directly by an engineer.
- The tool layer comprises Synopsys EDA and simulation and analysis ground-truth engines that agents use to perform specific functions and validate outcomes. These engines execute the requested work but do not establish goals or make decisions.
Engineering teams can establish checkpoints where people inspect results, validate decisions, and redirect the workflow. Teams can begin with frequent review and reduce intervention as they gain confidence in the outcomes.
Open by design
Engineering environments are heterogeneous by nature. Companies have their own infrastructure, models, data, agents, tools, workflows, and ecosystem integrations. Autonomous capabilities need to operate within those environments without requiring customers to replace what already works.
The Synopsys Autopilot Platform is designed to provide that flexibility. Customers can combine Synopsys and third-party agents in the same workflows, integrate their own engineering data, and choose among commercial, open-source, and fine-tuned language models based on their requirements. The platform also supports Synopsys-managed and customer-managed deployments, including Synopsys Cloud, customers’ cloud environments, and on-premises infrastructure.
Security and data management are part of the platform foundation. Architecture flexibility, access controls, encryption, and runtime guardrails help protect customer, partner, and Synopsys intellectual property while giving engineering organizations control over how autonomous workflows are deployed.
Demonstrated engineering outcomes
More than 50 customer engagements have demonstrated measurable gains across several areas:
- Engineering productivity reflects the time and manual effort required to complete engineering work.
- Design quality includes outcomes such as greater verification coverage and improved power, performance, and area (PPA).
- Workflow closure measures how quickly teams can reach verification, implementation, and other engineering objectives.
- Efficiency accounts for token usage and latency involved in completing the work.
Demonstrated results include up to 50× faster verification closure, 20% higher coverage, 2× better token efficiency, and lower latency. In one customer engagement, Synopsys Verification AgentEngineer capabilities also delivered a 10% to 30% productivity improvement in RTL code generation through more efficient generation of SystemVerilog assertions, wrapper modules, and parameterized modules, as well as code refactoring.
These results show how long-horizon agents and context intelligence can help engineering teams complete workflows faster, improve design outcomes, and use model resources more efficiently.
Advancing autonomous engineering from silicon to systems
Designing advanced chips and systems requires specialized expertise, long-horizon orchestration, and continuous judgment spanning interconnected activities, tools, and disciplines.
Built on the Synopsys Autopilot Platform, Synopsys AgentEngineer solutions apply those capabilities to complex engineering objectives. They reason through the work, coordinate specialized execution, respond to intermediate results, and adapt their plans while giving organizations control over the level of human oversight.
The result is faster turnaround, stronger engineering outcomes, and more efficient execution across silicon-to-systems workflows.
Continue Reading
Agentic AI in Chip Design: Why Knowledge Becomes the Advantage (Part 3 of 4)
Agentic AI in Chip Design: The Rise of Autonomous Engineering (Part 2 of 4)