Verisium AI-Driven Verification Platform

Verisium Debug

Comprehensive debug solution from IP to SoC and from single-run to multi-run, offering fast interactive and post-process debug with waveform, schematic, driver tracing, and SmartLog technologies. Verisium Debug is natively integrated with the Cadence JedAI Platform and other Verisium apps to enable AI-driven root cause analysis of bugs with the support of simultaneous automatic comparison of passing and failing tests.

Learn More

Verisium AutoTriage

Builds AI models that help automate the repetitive task of regression failure triage by predicting and classifying test failures with common root causes.

Verisium CodeMiner

Provides an algorithmic solution to compare multiple source code revisions of an IP or SoC, classify these revisions, and rank which updates are most disruptive to the system's behavior to help pinpoint potential bug hotspots.

Verisium WaveMiner

Applies powerful AI engines to analyze waveforms across a full verification testsuite and determine which signals, at which times, are most likely to represent the root cause of test failures.

Verisium PinDown

Integrates with the Cadence JedAI Platform and industry-standard revision control systems to build AI models of source code changes, test reports and log files to predict which source code check-ins are most likely to have introduced which failures.

Verisium SimAI

The Verisium SimAI App harnesses the power of machine learning technology. It builds models from regressions run in the Xcelium simulator, enabling the generation of new regressions with specific targets. This includes efficient soak testing of the entire design or specific areas and improved regression efficiency. The Verisium SimAI app also features cousin bug hunting, a unique capability that uses information from difficult-to-hit failures to expose cousin bugs. With these advanced machine learning techniques, Verisium SimAI offers the potential for a significant boost in productivity, promising an exciting future for our users.

Verisium SpecMiner

The Verisium SpecMiner vision is to accelerate design verification by using advances in large language models to enable the processing of natural language specifications. This includes translating natural language into SVAs and extracting testable items from sections of a spec to form a vPlan.

These capabilities are integrated with existing Cadence tools to bring the benefits directly to the user. To minimize the complexity of providing GPU resources to the application, Verisium SpecMiner uses the Cadence JedAI Platform to connect to an LLM.

Verisium SmartRun

Verisium SmartRun aims to help users optimize regression turnaround time and improve verification throughput. It does this by analyzing historical data from regressions and providing reports that allow users to determine their preferred queuing policy and choose settings that will achieve the user’s turnaround objectives by reshuffling the order of runs in their regression.

Verisium AutoFocus

Verisium AutoFocus is a tool that analyzes tests, designs, and design changes and recommends the most appropriate set of tests to verify those changes. It enables users to develop smarter, nimbler, more targeted regressions and helps to reduce failure escapes from one level of verification to the next, in addition to helping users make better use of available compute resources.

Jasper FormalAI

Jasper FormalAI uses advanced learning techniques to achieve better proof convergence in less time. Jasper FormalAI uses multiple different techniques at various stages to orchestrate its many proof solvers to achieve improved results. First, Jasper FormalAI uses reinforcement learning to select and parameterize solvers to enable faster first-time proofs. Second, Jasper FormalAI uses machine learning to optimize successive runs for regression testing on-premises or in the cloud. The goal is not just to re-run the proofs to get the same results on a new version of the design-under-test (DUT) faster but to learn from previous runs to obtain more proofs, find more counterexamples, and achieve deeper bounds on the hard-to-converge proofs. Jasper FormalAI users typically see a 2X to 4X increase in proof performance out-of-the-box and a 5X to 10X improvement on regression runs.