The Gemini Live Agent Challenge has concluded, celebrating the creativity and technical prowess of developers from around the globe. With participation from 11,878 developers and 1,536 projects submitted across 151 countries, the challenge aimed to advance the integration of multimodal capabilities in AI agents.
Participants were tasked with creating agents that could see, hear, speak, and create in real time using the Gemini Live API and Google Cloud infrastructure. The entries were evaluated in three categories: The Live Agent, The Creative Storyteller, and The UI Navigator.
Celebrating Winners at Google Cloud Next ‘26
Among the winners, Jeremiah Somoine and Bryen Param were honored at Google Cloud Next 2026 in Las Vegas, where they shared insights and experiences with the developer community. Both presented Lightning Talks and participated in interviews, discussing their innovative projects.
Bryen's project, drone-copilot, focuses on enabling natural conversations with drones, while Jeremiah's Sankofa transforms family histories into immersive narratives through AI storytelling. Their experiences highlight the importance of hands-on engagement with technology for aspiring developers.
List of Winners
| Category | Project Name | Developer(s) | Description |
|---|---|---|---|
| Grand Prize | ORION | Aditya Shukla | A voice-directed surgical co-pilot for robotic surgery providing real-time assistance. |
| The Live Agent | drone-copilot | Bryen Param | Allows users to control drones through natural speech, enhancing interaction. |
| Creative Storyteller | Sankofa | Jeremiah Somoine | A multimodal AI storyteller that creates narratives from user details. |
| UI Navigator | Moonwalk | Enaiho Uwas Paul, Aman Kumar Sah | A hands-free assistant for intuitive computer navigation using voice commands. |
| Best Multimodal Integration | Wand | David Li | A browser assistant that uses speech and gestures for seamless navigation. |
| Best Technical Execution | JohnKeats.AI | Matthew Keats | An emotional companion that listens and responds to users' emotional cues. |
| Best Innovation | Rayan Memory | Yusuf Elnady | A 3D memory palace that helps users explore their memories interactively. |
Honorable Mentions
- NagarDrishti: A voice assistant for reporting road conditions.
- Ekaette: A multimodal assistant for customer service across various platforms.
- VibeCat: A proactive desktop companion that anticipates user needs.
- Call My Parts: Automates sourcing used vehicle parts through voice commands.
- Relay: An interactive lab partner for electronics projects.
Next Steps: Developers are encouraged to join initiatives like the Gemini Enterprise Agent Ready (GEAR) program to further their skills and contribute to the evolving landscape of AI technology.
![[Weekly funding roundup May 9-15] Rapido deal lifts the VC inflow](https://cdn.wtguru.com/news/thumbnails/19788.jpg)