Google Vertex AI Agents¶
Vertex AI External Agents allow you to make a Vertex AI Agent available as a standard Dataiku Agent
Once configured, it can be leveraged anywhere a Dataiku Agent is supported, including:
The Dataiku Flow (notably Prompt Recipes)
Chat UIs such as Agent Hub
The LLM Mesh API
/agents/agent-review
/agent/agent-evaluation
It supports streaming responses, and is compatible with agents built using both Langchain and Google ADK frameworks in Vertex
Prerequisites¶
Vertex AI support is provided by the “Google Vertex AI Agents” plugin, which you need to install
You need a Vertex AI LLM Connection configured, and an existing Vertex AI Agent deployed using either the Langchain or ADK framework.
Usage¶
Navigate to your Dataiku project.
Go to Agents & GenAI Models > + New Agent.
Select Vertex AI - Agents and provide a unique name.
Connection: Select your pre-configured Vertex AI LLM Connection.
Agent Framework: Choose Langchain or Google ADK.
Agent Name: Select your agent from the dropdown list.
You can then use the built-in chat interface within the Agent configuration page to test your agent.
Cautions¶
In order to properly support multi-turn conversation, when using the LLM Mesh API, and in ADK mode, you need to pass a stable
conversationIdin the request context. Note that when using the built-in chat, or Agent Hub, this is automatically handledOnly text-only interactions are supported. Image inputs are not supported