Build reliable multi-agent applications with ADK Go 2.0. Discover our new graph-based workflow engine, built-in human-in-the-loop, and dynamic orchestration

2026-06-30 · Source: Google Developers Blog - AI · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, medium

Summary

ADK for Go 2.0, released on June 30, 2026, introduces a graph-based workflow engine for building reliable multi-agent applications in Go. This update allows developers to compose complex orchestrations, including classification, branching, fan-out, and looping, using a clean, idiomatic API. Key features include built-in human-in-the-loop (HITL) capabilities, dynamic orchestration written in plain Go, and new LLM agent modes (Chat, Task, SingleTurn). The unified node runtime ensures single agents and full graphs operate on the same execution model, enhancing resilience with per-node retry policies and timeouts. This release builds upon ADK 1.0's foundation, offering a robust framework for production-grade agent workflows.

Key takeaway

For AI Engineers building complex Go-based agent applications, ADK for Go 2.0 significantly simplifies orchestration challenges. You should adopt its graph-based workflow engine to define robust, observable, and resumable multi-agent systems. This update allows you to integrate human approvals and dynamic logic directly, reducing brittle ad-hoc control flow and improving overall system resilience. Explore the new workflow examples to quickly implement these advanced capabilities.

Key insights

ADK for Go 2.0 enables reliable multi-agent workflows through a graph-based engine, integrating HITL and dynamic orchestration.

Principles

• Graph-based workflows enhance reliability.
• Nodes abstract diverse units of work.
• Edges enable flexible control flow.

Method

Define application shape as a graph of nodes connected by edges. Use typed node constructors for functions, agents, and tools. Implement routing conditions on edges for control flow. Orchestrate dynamically with RunNode(...) calls.

In practice

• Chain nodes using workflow.Chain.
• Embed sub-workflows as single nodes.

Topics

• ADK Go 2.0
• Multi-Agent Workflows
• Graph-based Orchestration
• Human-in-the-Loop
• Go Development
• LLM Agent Modes

Code references

• google/adk-go

Best for: AI Architect, AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• From Dashboards to Decisions: How We’re Building AI Agents at RaySuite AI — AI agents transform user interaction by providing intelligent, contextual answers and taking action over complex systems.
• Building Production-Ready AI Agents with Agent Development Kit — ADK simplifies building scalable, production-ready multi-agent AI systems by abstracting orchestration and integrating cloud services.
• TAI #207: Claude Opus 4.8 Is Better, but Dynamic Workflows Are the Bigger Story — Dynamic Workflows allow LLMs to self-orchestrate complex tasks across many subagents, enhancing performance and efficiency.
• Building Human-In-The-Loop Agentic Workflows — Human-in-the-loop design is critical for LLM agents to mitigate errors and ensure quality in subjective tasks.
• Google ADK for Java 1.0 Introduces New App and Plugin Architecture, External Tools Support, and More — ADK for Java 1.0 enhances AI agent development with new tools, architecture, and collaboration features.
• Workflow Mapping For Agentic Systems & Knowledge Graphs — Complex, multi-faceted workflows are better managed with knowledge graphs than traditional prompt-based methods.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Google Developers Blog - AI.