Supercharge your AI agents: Inside the new ADK integrations ecosystem
summary
Demo shows a concrete end-to-end pipeline: ADK spins up a multi-agent summarizer+translator in less than two minutes, the Hugging Face → GitHub → Notion “research scout” runs from ~60 lines of code and produces a Notion briefing, and a Daytona + Chroma setup executes code in a safe Daytona sandbox while storing snippets in memory. The ecosystem uses ADK Dev Skills and MCP docs, supports Python/TS/Go/Java, and the demos complete successfully.
AI-written summary · about
chapters (19)
- 0:00Welcome
- 1:00Agent Development Kit
- 3:57Beyond Hello World prototypes
- 5:50Agents with external integrations
- 7:41Coding with ADK Dev Skills
- 9:32ADK in IDEs like Antigravity
- 13:26Live multi-agent demo
- 18:52ADK web interface features
- 22:52ADK integrations ecosystem
- 26:50Research scout with Hugging Face GitHub Notion
- 35:24Workflow automation with Cron
- 37:13Plugins versus MCP tools
- 40:52Coding agent with Chroma Daytona
- 44:47Multi-agent podcast workflow
- 48:43Recap of demos and key takeaways
- 50:30Q&A on subagents
- 54:25Wrapping APIs as tools
- 57:37State management at scale
- 1:01:15Closing
related talks
01h 09mMulti Agent Architecture using ADK
54mCreate Multi-Agent Systems with ADK
45mUnifying AI experience for customers: Building multi-agent systems with Google ADK
07h 44mAI Agents Conference 2026 - Day 1
07h 47mAI Agents Conference 2026 - Day 5
19mBuilding Agents at Cloud Scale
06mKeynote: Building Agents at Scale
01h 08mFrom Model Integration to Agents: Building Intelligent .NET Applications