How to build a custom vision agent
Google Cloud Tech
Watch on YouTube ↗ · 5:50Description.
Join Gemini Enterprise Agent Ready (GEAR) for the latest agent resources → https://goo.gle/4xpCT5W John's GitHub repo → https://goo.gle/4uuvsYi
Ever wanted to turn your webcam into a generative media engine? Google Developer Expert, John Capobianco shares how he created a custom vision agent using the Google Gemini ecosystem and MCP, capturing live photos with Nano Banana Pro and animating them using Veo 3. It even processes natural language prompts and real-time ASL conversations.
Speakers: John Capobianco Products Mentioned: Gemini, Model Context Protocol, Nano Banana, Veo 3
Original description by Google Cloud Tech. View the source on YouTube ↗
Transcript.
Caption tracks are available on the original video. Open it on YouTube, expand the description, and choose Show transcript when offered.
Languages: en. Some captions are automatically generated and may contain errors.
Open YouTube for transcript ↗More conversations.
In this episode, John Capobianco of Itential takes us through a more practical AI journey.
Sep 29, 2026 · 27:14
CREATE TRAVERSALCREATE TRAVERSAL and ORF-RE SITE-CAV, featuring John Capobianco
Sep 24, 2026 · 1:14:13
TFiRStop Shadow AI Before It Owns Your Network | John Capobianco, Itential
Sep 16, 2026 · 17:54