NVIDIA Developer

Seattle DGX Spark Hackathon Winners Spotlight

Published 2026-09-11 · Duration 40:58

Summary

This summary covers the NVIDIA DGX Spark Hackathon winners, spotlighting two advanced local AI applications: Kerberos, a shared spatial-awareness system for search-and-rescue (SAR) teams, and VELA, a voice-first, consent-controlled healthcare action system. Both projects demonstrate the power of running complex, multi-agent AI workflows entirely on local hardware (NVIDIA GB10), ensuring data privacy and real-time action capability.

Download summary

Key takeaways

  1. Local AI for Privacy and Reliability 22:40

    Both winning systems (Kerberos and VELA) are designed to run entirely on the NVIDIA GB10, ensuring that sensitive data (e.g., medical records, live camera feeds) remains local and is not transmitted to the cloud, addressing critical security concerns in healthcare and SAR.

  2. Agentic Workflows for Complex Tasks 25:40

    The projects utilize multi-agent architectures (e.g., VELA's system) where specialized agents (like Parakeet for speech recognition, Neatron for reasoning, and Magpie for speech output) collaborate to perform complex, multi-step tasks, moving beyond simple chatbots to actionable outcomes.

  3. Shared Situational Awareness in SAR 3:40

    Kerberos creates a shared live map for SAR, integrating data from multiple sources (drones, robots, body cameras) to track responders, map searched areas, and pinpoint casualties or hazards, even indoors where GPS fails.

Technical details

  • Hardware & Architecture 300s

    The applications are built to run locally on the Acer Veriton GN100, powered by the NVIDIA GB10 Grace Blackwell Superchip, emphasizing edge computing and local processing.

  • AI Models and Algorithms 380s

    The systems leverage advanced models including Neatron (used in both projects) for reasoning and multi-modal understanding, and incorporate algorithms like DPVO and SLAM for accurate odometry and mapping.

  • Healthcare Workflow (VELA) 1540s

    VELA processes user care needs and insurance information to generate verified options across cost, coverage, and providers. It uses a multi-agent structure with Parakeet (speech recognition), Neatron (reasoning), and Magpie (spoken response).

  • Security and Control 1750s

    VELA implements 'explicit consent' before taking any consequential action. Furthermore, the system uses a sandbox environment and OpenShell to define boundaries, ensuring agents operate within controlled policies.

Mentioned resources

  • NVIDIA GB10 Grace Blackwell Superchip (Hardware)
  • Isaac Sim (Simulation Platform)
  • Neatron (AI Model/Agent)
  • Parakeet & Magpie (Speech Models)

Channel & topics

Watch on YouTube · Back to latest

This independent, AI-assisted summary is provided for commentary and informational purposes. It may contain errors or omit important context. Please watch the original video for the creator's complete presentation. Video, thumbnail, and related copyrights belong to their respective owners.