Build Visual AI Agents From a Prompt With NVIDIA Cosmos and VSS 3.3

Building visual AI agents requires connecting live video streams alerts, search, and insight—and optimizing the cost of AI processing on hours of footage. Join us live to build a visual AI agent from a natural-language prompt using NVIDIA Cosmos and NVIDIA VSS 3.3. See how the Build Vision Agent skill helps reduce development effort and how Adaptive Efficient Video Sampling (EVS) saves on video processing costs. You’ll learn how to: - Use a coding agent and VSS skills to combine alert verification, video search, summarization, and reporting. - Review and validate the deployment, reuse shared services, and add capabilities to a running application. - Configure Adaptive EVS and measure its impact on token usage, throughput, and latency. Bring your questions and follow the workflow from prompt to deployment. Resources: VSS 3.3 Tech Blog: https://nvda.ws/4e1zxOf VSS github: https://nvda.ws/479uZBP VSS Build: https://nvda.ws/4wnM71T Brev Launchable: https://nvda.ws/476yr08 VSS Documentation: https://nvda.ws/4nnuj2M VSS Github: https://nvda.ws/4woAJmy