Synopsys Launches AI Agent Platform to Automate Chip Design with 50+ Industry Engagements

Synopsys Debuts Autopilot Platform for Autonomous Chip Development

Synopsys announced its Synopsys Autopilot Platform and Synopsys AgentEngineer solutions, introducing what the company describes as the industry's broadest portfolio of domain-specific long-horizon AI agents designed to automate engineering workflows across semiconductor and systems design1

1

2

2

. The AI agent platform covers six primary domains: verification, system validation, implementation, analog and mixed-signal design, manufacturing, and simulation and analysis3

3

. More than 50 customer engagements are currently underway, with general availability planned for the end of 20261

1

.

Architecture Enables AI-Driven Autonomous Engineering

The Synopsys Autopilot Platform provides orchestration, skills, memory, telemetry, and governance for autonomous engineering workflows, creating what Synopsys calls a trusted environment for AI systems to coordinate activities and execute actions2

2

. Anand Thiruvengadam, executive director of product management at Synopsys, detailed a three-layer architecture: long-horizon AgentEngineers function as domain-specific super agents that orchestrate task agents, task agents complete specific bounded engineering tasks, and the tool layer's engines execute requested work without setting goals or making decisions1

1

. Unlike long-running agents that perform one activity for extended periods, these domain-specific long-horizon AI agents address goal complexity, pursuing objectives requiring hundreds or thousands of reasoning steps1

1

.

Platform Delivers Flexibility and Context Intelligence

The platform's cognitive model powers what Synopsys terms context intelligence, combining deep domain engineering knowledge, reusable skills, and persistent memory to help AI agents interpret design intent and guide workflows3

3

. Customers can bring their own language models, data, and infrastructure, choosing between commercial, open-source, or fine-tuned models while deploying on Synopsys Cloud, their own cloud, or on-premises infrastructure1

1

. Access controls, encryption, and runtime guardrails protect customer, partner, and Synopsys intellectual property, particularly important when third-party agents share workflows1

1

.

Performance Gains Demonstrated Across Verification and Implementation

Synopsys reports performance improvements including up to 50x faster verification closure, 20% higher coverage, 30% productivity boost, 2x better token efficiency, and reduced latency1

1

. The 50x faster verification and 20% coverage figures stem from July work with Nvidia, measured against Synopsys' own verification workflows without AgentEngineer1

1

. Fujitsu reported a 10% to 30% productivity increase in RTL code generation, with the 30% figure representing the top of this range1

1

2

2

. The 2x token efficiency improvement came from an unnamed customer comparing Synopsys' agents with its own agents built on commercial agentic harnesses1

1

.

Industry Leaders Validate Autonomous Chip Development Approach

Intel, Samsung, MediaTek, TSMC, and Nvidia have endorsed the technology through active engagements2

2

. Intel described work with the Verification AgentEngineer technology for debug processes, while MediaTek referenced collaboration on an Agentic AI AMS Platform for analog and mixed-signal design2

2

. Samsung Electronics noted the solutions assist engineering teams with development workflows for advanced memory technologies2

2

. TSMC mentioned AI-based solutions for power integrity analysis workflows in multi-die 3DIC assembly architectures2

2

. AheadComputing's Vice President of Verification, Alon Mahl, stated the Implementation AgentEngineer helped reduce manual engineering effort from RTL handoff through signoff1

1

.

Autonomous Agents Operate With Human Oversight

Ravi Subramanian, chief product management officer at Synopsys, stated the technology allows companies to accelerate their shift from AI-assisted design to autonomous engineering1

1

. While Synopsys describes the agents as autonomous, approval checkpoints remain with humans1

1

. The verification loop demonstrates how agents plan, orchestrate task agents, check returned results, and adjust when intermediate results fall short1

1

. When tests find bugs, root-cause analysis agents read logs, cluster errors, form hypotheses, inspect waveforms for confirmation, make local rewrites of RTL to prove fixes, and produce bug fix manifests1

1

. This launch follows demonstrations of agentic AI workflows developed with Nvidia and Microsoft at the Design Automation Conference in July1

1

.