Vertex AI Agent Builder Example

 Vertex AI Agent Builder Example

Platform: Vertex AI Agent Builder (playbook-based)

Framework: RAG (Retrieval-Augmented Generation)

System Architecture

The system uses a Retrieval-Augmented Generation (RAG) architecture to bridge Large Language Models (LLMs) with private financial datasets

Core Engine: Gemini 2.5 Flash (Optimized for low-latency reasoning and table parsing).

Orchestration: Playbook-based Vertex AI Agent Builder

Knowledge Base: Unstructured Data Store (Semantic Index) connected to Google Cloud Storage.

Go to Builder Click link below

Vertex AI Agent Builder

The API needs to be enabled

Click Continue and activate API - 

I clicked Allow Google to sample model

Click Create App

Click Conversational agents

Then click Create

Options for creating an agent

Select Build your own

Enter a Display Name - Finance Intelligence Agent

Location - global

Time Zone - (GMT 5:00) America/New_York

Leave Playbook checked and hit Create

For Display name - Finance Analysis Agent

Goal -Help users analyze stock performance and retrieve financial metrics from SEC annual reports

Instructions - Greet the user, then ask which stock they want to analyze today

Click Save

Using the 3 dots in the upper right hand corner of your screen select Toggle simulator

Environment - Draft

Start Resource - Finance Analysis Agent

Models - gemini-2.5-flash

Start a conversation with the agent. 

It won’t provide much information because we have not connected any data sources to it.

Go back to the Conversational Agents screen area and click +Data store

Tool Name - Finance Reports

Type - Data store

Description - Tool to retrieve financial metrics (Revenue, Net Income, Operating Expenses) for Stocks e.g. Apple, Google, and Microsoft

Then hit Save

Next Click on Create data store

Click Cloud Storage (unstructured data)

Data store name - Finance Annual Reports

Synchronization frequency - One time

Select a folder or a file you want to import - Use browse to select your cloud storage bucket

Hit Create

It will take a long time for this process

Next add data stores

Select the data source hit Confirm

Hit Save

Go To Playbook Basic page

Update your instructions - Use ${TOOL: Finance Reports} for any request regarding Annual Metrics (, Microsoft, or Google's financial reportings.”),

 Click SAVE

Break down Apple's revenue across different products and regions of the world

Or any other company loaded