Founder Kevin McNamara goes deeper on product and customers as the company scales to meet accelerating demand. Zack Novak enters as CEO.

If you are building a Physical AI system, you have probably discovered an uncomfortable truth: your real-world sensor data is never as clean as the papers assume. We exist to turn our messy reality into a true “Parallel Domain": a digital reconstruction of reality that can be trusted to test and validate autonomous systems. Customers across automotive, drone delivery, eVTOL, agTech, last-mile delivery, and robotics trust us, and NVIDIA is an enabling partner key to improving our product and operating at scale.

The industry faces a stark reality: developing safe autonomous systems requires billions of test miles, yet a single failure can set back years of progress.

We are excited to unveil our latest technical advancement at CES: the integration of NVIDIA Omniverse NuRec Fixer into our PD Replica pipeline.

Joint solution accelerating AI-powered AV development, validation, and testing

"Shift Left" is a transformative paradigm that reorients the entire testing and validation process. Instead of waiting until the later stages of development to test on real-world tracks and roads, it advocates for integrating testing through advanced simulation, from the very beginning and continuously through the development cycle. Think of it like moving quality control from the end of the assembly line to every single step of the manufacturing process.

Parallel Domain today announced the completion of a real-world fidelity Replica of the Mcity Test Facility at the University of Michigan. This real-fidelity simulation environment allows Mcity members and researchers to run perception, planning, prediction, and control tests with realism nearly indistinguishable from the physical track.

Pedestrian protection is no longer optional. It’s becoming a regulatory requirement, and perception systems must be tested rigorously to ensure they can perform reliably in the real world. But how can teams meet this demand at scale? Real-world testing is expensive, slow, and often fails to capture rare but critical edge cases. Simulation provides a powerful alternative, but only if it can be trusted. In this blog post, we put PD Replica Sim to the test by benchmarking it against real-world data from the Waymo Open Dataset.

PD Replica employs advanced scene reconstruction and rendering techniques to provide developers with realistic simulation environments suitable for open and closed-loop testing and validation of perception models. However, while PD Replica Sim can programmatically generate scene variations, there are limitations to the full breadth of environmental controls available today, and where Cosmos Transfer with an open-ended prompt can enhance the quantity of variations available.

ML teams can use PD Replica Sim to push their autonomous system to its limits, interrogating model weaknesses and testing its generalization more rigorously than real-world data typically allows —covering tricky edge cases, handling occlusion and orientation sensitivity, and even evaluating model performance on unseen traffic sign types.