Google Released Agent Development Kit for Kotlin

The new framework brings production-ready AI agent building capabilities to the Kotlin ecosystem.

Updated on Sept. 20, 2026 in Artificial Intelligence

Bold vector editorial illustration of a modular server processor with metallic pins, representing technical software development frameworks.
Google has launched its Agent Development Kit for Kotlin, an open-source framework enabling developers to build multi-agent systems with enhanced type-safety. AI Illustration. Upload story photo >

Live Poll

Would you prefer to build AI agents using native language frameworks like Kotlin?

Google has launched the Agent Development Kit for Kotlin 1.0, an open-source framework designed for constructing multi-agent systems. The tool now offers feature parity with existing Google ADK versions for Python and Java.

Why it matters

The release allows developers to leverage compile-time function schema generation to enhance type-safety and performance. It also introduces critical human-in-the-loop workflows to manage sensitive tool operations safely.

The ADK 1.0 framework supports Android and JVM server applications through Kotlin Multiplatform. It integrates with LiteRT-LM and ML Kit for on-device inference alongside Firebase AI Logic for cloud capabilities.

The players

Google

Google is a global technology company that develops software, hardware, and artificial intelligence solutions.

The details

Built on Kotlin Multiplatform, the toolkit enables developers to serialize, pause, and restore agent states through built-in session management. It utilizes Kotlin Symbol Processing to avoid runtime reflection, while domain-specific knowledge can be loaded dynamically via agent skills.

Timeline

  1. Google released the Agent Development Kit for Kotlin 1.0 on September 20, 2026.

The Tech Race

This release signals a transition toward unifying AI development frameworks across major programming languages. By extending the Google ADK to Kotlin, the company is competing to standardize agent building within the JVM and Android mobile ecosystems.

Developers can now build AI agents with improved performance and stronger type-safety within their existing Kotlin codebases. These updates facilitate easier on-device inference and more reliable cloud-based logic integrations for end-user applications.

The takeaway

The adoption of Kotlin Multiplatform for this ADK demonstrates a broader industry shift toward cross-platform AI development. Developers should prioritize the new human-in-the-loop features to ensure secure AI interactions in their applications.

Further reading

For more information on the evolving tools used to build modern systems, visit Artificial Intelligence.

More information

View the source code or contribute to the project at the ADK for Kotlin GitHub repository.

Live Poll

Would you prefer to build AI agents using native language frameworks like Kotlin?