Real Intent, announced the launch of its Riven Sign-Off AI Agent on July 23, 2026, in Sunnyvale, California, designed to help design and verification engineers reach sign-off up to ten times faster across all of Real Intent's static sign-off tool workflows. Riven uses agentic and generative AI to speed up product knowledge access, error analysis, and automated debugging.
Static sign-off is widely used in system and semiconductor design to enable earlier RTL and SoC-level verification. As designs grow more complex, an increasing number of application areas now require static sign-off, including RTL linting, clock domain crossing, reset domain crossing, design for test, connectivity and glitch analysis, and hardware security. Real Intent's new Riven agent significantly cuts engineering effort by speeding up static sign-off review and closure by as much as 10 times. It does this by applying generative and agentic AI to help engineers automate Real Intent's sign-off product workflows, analyzing design failures using its deep product and workflow expertise.
Riven supports several practical use cases. Engineers can use it to quickly get technical guidance, best practices, and explanations of error messages, without digging through extensive documentation or needing deep command-level familiarity with each product. They can also direct Riven to their workspace and have it review the design database to identify the most critical violations in a block or SoC, guiding them to the highest-priority issues. In addition, engineers can ask Riven to trace the root cause of a specific error and suggest a possible fix, which they can then accept or modify before rerunning the sign-off tool.
Several additional features strengthen Riven's capabilities. It builds on Real Intent's established, proven violation analysis across its supported domains: RTL linting, microarchitectural integrity, clock domain crossing, reset domain crossing, design for test, connectivity and glitch, and hardware security. Customers and partners can also extend Riven's functionality by integrating their own custom LLM agents and workflows through a standard Model Context Protocol (MCP) interface. Riven further works with popular IDEs, including VS Code.
