Linker Vision Highlights Video Reasoning AI at NVIDIA GTC 2026

San Jose, CA – March 16, 2026 – Linker Vision, a leading AI platform company pioneering Physical AI and Reasoning AI, is showcasing its Video Reasoning AI platform at NVIDIA GTC 2026, designed to power intelligent operations across smart cities and smart spaces. Deployed in Taiwan and Vietnam, Linker Vision is working with Voxelmaps and Inchor for digital twin traffic simulation capabilities to test the possibility of city operation agents that can perceive, simulate, reason, and act in the City of San Jose. 

The platform enables contextual understanding of live video streams, supporting situation assessment, impact analysis, and decision support across traffic operations, industrial monitoring, infrastructure management, and public safety. Leveraging the NVIDIA Metropolis Blueprint for video search and summarization (VSS) for large-scale video analytics and NVIDIA Cosmos™ open world foundation models (WFMs) for world understanding and reasoning, Linker Vision transforms visual input into structured, actionable agentic intelligence. 

At the core is a digital twin–enabled training pipeline aligned with the NVIDIA Physical AI Data Factory Blueprint. The Physical AI Data Factory Blueprint serves as a single reference architecture that unifies data curation, generation, and validation at scale. This blueprint uses Cosmos  WFMs including Cosmos Transfer and Cosmos Evaluator, powered by Cosmos Reason, to transform limited training data into large, diverse datasets. Linker Vision is actively testing a Physical AI toolchain on Azure that integrates the blueprint with several Azure services. This approach supports scalable model development and continuous improvement for complex physical environments, producing model-ready training datasets from raw data.

To process high‑volume, real‑time video streams at scale, Linker Vision is collaborating with telecom operators who are building AI grids—geographically distributed and connected AI infrastructure—based on NVIDIA's reference design.

With AT&T, the system runs on edge sites that act as AI grid nodes, bringing video reasoning closer to cameras and sensors for lower latency and more reliable operations.

With T‑Mobile, Linker Vision is piloting distributed deployments powered by a grid of AI‑RAN ready infrastructure powered by NVIDIA RTX 6000 PRO Blackwell Server Edition. An active smart city project in San Jose integrates point cloud data, 3D digital twins, and vision AI for real-time traffic optimization and urban planning.

Building on this foundation, Linker Vision connects simulation, training, and deployment into a continuous AI lifecycle, bringing Physical AI and Reasoning AI into real-world operations. 

“As digital twins operate everywhere, reasoning AI can scale with them—unlocking new opportunities for smart cities and critical infrastructure," said Paul Shieh, Founder and CEO of Linker Vision.