Google ADK livestream builds Python agents

// 1h agoVIDEO

Google ADK livestream builds Python agents

Google Cloud Tech has announced a livestream tutorial showcasing Google's Agent Development Kit (ADK) and Gemini 3.5 Flash. The livestream will walk developers through the process of building a Python-based AI agent to conquer "The Garden of Forgotten Prompts," highlighting how to design, coordinate, and run multi-agent workflows using Google's open-source framework and LLM tooling.

// ANALYSIS

Google is actively pushing its open-source Agent Development Kit (ADK) to compete with established orchestration libraries like LangGraph by highlighting practical developer tutorials and live builds.

–Demonstrating the toolkit with a game-like task ("The Garden of Forgotten Prompts") is an effective dev-rel strategy to display agentic reasoning, memory, and tool usage.
–This livestream highlights Google's focus on Gemini 3.5 Flash's speed, cost-efficiency, and tool-calling capabilities when orchestrated via ADK.
–Increasing educational content around ADK aims to lower the barrier to entry for building complex, production-ready multi-agent architectures on Google Cloud.

// TAGS

google-adkgeminipythonagentdevtooltutoriallivestream

DISCOVERED

1h ago

2026-06-17

PUBLISHED

1h ago

2026-06-17

RELEVANCE

7/ 10

AUTHOR

GoogleCloudTech

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE21m ago

Electric showcases collaborative audio Yjs documents

Electric co-founder Sam Willis demonstrated a collaborative editing environment combining Yjs document synchronization with real-time audio and multiple cooperative AI agents. By treating agents as first-class CRDT peers, the system displays their real-time cursors and selections using Electric's local-first sync and Durable Streams for orchestration and state persistence.

NEWS24m ago

Higgsfield drops unlimited Seedance video-gen

Higgsfield has announced a promotional campaign offering a full month of unlimited video generation using its Seedance AI model. Described as the beginning of the "token-maxxing era," this initiative allows creators to generate high-quality, cinematic AI videos without standard credit or token limitations, driving user adoption and platform engagement.

INFRA35m ago

LM Studio clusters Macs for 1T model

During WWDC, LM Studio demonstrated a collaboration with Apple that ran the 1-trillion-parameter Kimi K2.6 model locally across a cluster of four Mac Studios using a preview version of LM Studio. They also demonstrated secure remote access to this private cluster from a MacBook Neo and an iPhone using LM Link.