[Video](https://www.youtube.com/watch?v=jinUBmblW5k)

*Customer Experience Agent Studio* (CX Agent Studio)
is a minimal code conversational agent builder.
It uses AI and an agent builder user interface
to guide users through the process of creating agent applications.
It also uses AI to have controlled conversations with end-users at runtime.

CX Agent Studio is built on
[Agent Development Kit (ADK)](https://google.github.io/adk-docs/)
and provides many simplifications, features, and customizations on top of ADK
that are tailored for enterprise use cases.
CX Agent Studio brings the power of ADK
to a broader audience without requiring coding expertise.

CX Agent Studio represents the evolution of Dialogflow CX
with the following advantages:

- [AI-Augmented building, optimization, and evaluation](https://docs.cloud.google.com/gemini-enterprise-cx/cx-agent-studio.md.txt#augmented) to accelerate development and improve quality.
- **Simple, intuitive visual builder** that empowers non-technical teams and agent architects.
- **Solves the "last mile" problem** by handling infrastructure, enterprise integrations, security, and operational complexities out-of-the-box.
- **Maintains natural conversational flow** during backend tool calls (truly async), eliminating awkward silences for the user.
- **Ultra-low latency voice interactions** through a bi-directional streaming architecture, optimized for natural-sounding conversations.
- **Integrated collaboration and versioning tools** offers in-platform features for team development, including granular changelogs, one-click rollbacks, and conflict warnings.

## AI-Augmented building, optimization, and evaluation

CX Agent Studio provides multiple features that use AI
to accelerate agent development and improve quality:

- [Create an agent with Gemini](https://docs.cloud.google.com/gemini-enterprise-cx/cx-agent-studio/quick/generate-agent.md.txt) to generate agent details for you.
- [Restructure instructions](https://docs.cloud.google.com/gemini-enterprise-cx/cx-agent-studio/instruction.md.txt#restructure) to reformat hand-written instructions in an XML structure, which can improve model instruction handling.
- [Refine instructions](https://docs.cloud.google.com/gemini-enterprise-cx/cx-agent-studio/instruction.md.txt#refine) to improve the clarity of your hand-written instructions.
- [Test case hill climbing](https://docs.cloud.google.com/gemini-enterprise-cx/cx-agent-studio/evaluation.md.txt#find-issues-with-ai) to improve the rigor and stability of your evaluations test cases.

## Legacy conversational agents

If you're looking for the documentation for legacy conversational agents, see:

- [Dialogflow CX](https://docs.cloud.google.com/dialogflow/cx/docs.md.txt)
- [Dialogflow ES](https://docs.cloud.google.com/dialogflow/es/docs.md.txt)