ADK
Also written: Agent Development Kit
Agent Development Kit (ADK) helps build and test AI conversation programs, using simulated users and AI voices.
The Agent Development Kit, or ADK, is a set of tools that helps developers create and test AI programs designed to have conversations. Think of it like a workshop for building chatbots or voice assistants.
For example, Google's ADK can be used to test how well a new voice assistant understands commands and responds naturally, even simulating different user voices and accents.
The Agent Development Kit (ADK) is a framework designed to streamline the development and evaluation of conversational AI agents. It typically provides components for dialogue management, state tracking, integration with large language models (LLMs), and tools for testing and simulation. This allows developers to build, debug, and deploy agents more efficiently.
ADKs often include features for synthetic data generation, automated testing with simulated user interactions, and performance metrics for evaluating agent behavior, such as task completion rates and response quality. Some ADKs may also integrate with text-to-speech (TTS) and speech-to-text (STT) engines for voice-based agent development.
Google is highlighting its Agent Development Kit (ADK) by adding new features for evaluating live voice agents. This focus on agent development tools suggests a growing emphasis on creating more sophisticated and interactive AI applications.
- How to Evaluate Live & Voice Agents in ADKGoogle · Google Developers Blog ·
- Google Gemini
- Google Gemini uses the Agent Development Kit (ADK) to build and test conversational AI agents, recently adding features for live voice agent evaluation.
The term 'Agent Development Kit' (ADK) is a descriptive name for a collection of tools focused on building AI agents. While specific implementations vary by company, the concept is a natural evolution in AI development tooling.
Explore the documentation for Google's ADK if available. Look for examples of how it simulates user interactions or integrates with AI models like Gemini TTS to create voice responses. Try to find a demo or tutorial that showcases building a simple conversational flow.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- Eval methodAutomated scoring of voice agent audio responses and tool executionsDefine evaluation scenarios and rubrics to automatically score voice agent audio responses and tool executions. · 15 min
- ToolGoogle Gemini for real-time voice AI agents with tool callingEnables voice AI agents to take actions mid-conversation without breaking the flow using the Agent Development Kit. · 30 min
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
Questions people ask
- What is the main purpose of an ADK?
- An ADK helps developers build, test, and improve AI programs designed to have conversations with users, like chatbots or voice assistants.
- How does an ADK differ from a regular software development kit (SDK)?
- An ADK is a specialized SDK focused specifically on creating and evaluating conversational AI agents, often including tools for dialogue management and user simulation.
First seen on Subvolts Sep 3, 2026. Updated Oct 19, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.