Build Anything with Gemini 3 - Agent Development and Multimodal AI Capabilities

Explore the groundbreaking capabilities of Gemini 3 Pro and Google's new Antigravity agentic development platform in this 28-minute podcast episode. Discover how Gemini 3 Pro achieves state-of-the-art reasoning and multimodal capabilities, topping the LMArena leaderboard with a breakthrough score of 1501 Elo and demonstrating PhD-level reasoning on advanced benchmarks. Learn to leverage Gemini 3 in Google AI Studio, utilize the annotation feature for enhanced development, and automate tasks using Gemini CLI. Master building multi-agent workforces, creating agents for ghostwriting and report automation, and developing video avatars using ADK, Gemini, and Veo 3.1. Gain practical insights into prompt engineering techniques for Veo 3.1 and understand how Antigravity is evolving the IDE experience toward an agent-first future with browser control capabilities and asynchronous interaction patterns. Follow along as hosts Smitha Kolan and Vlad Kolesnikov, joined by guest Brandon Hancock, demonstrate how to build anything from individual agents to complete agentic workforces using Google's latest AI technologies.