Build Your Personalized GenAI Search Engine with Vertex AI Search: A Step-by-Step Guide

Aadila Jasmin
Google Cloud for Startups
8 min readMay 9, 2024
An illustration highlighting the effortless search experience across diverse data sources using Vertex AI Search on Agent Builder

Is your company’s data a treasure trove of untapped insights? Are you tired of sifting through endless documents and spreadsheets, struggling to find the information you need? Imagine having a search engine as powerful as Google, but tailored specifically for your business data. That’s the magic of Google Cloud’s Vertex AI Search. Vertex AI Search transforms your raw data into actionable insights, creating personalized search experiences that make sense of website, structured and unstructured data. A solution that empowers you to make data-driven decisions with ease! In this blog, we will dive into the features, use-cases and a step-by-step guide to building your own Vertex AI Search Engine.

What is Vertex AI Search?

Vertex AI Search Engine, a part of Vertex AI Agent Builder, is a fully managed platform for developers to build Google-quality search experiences for websites, structured and unstructured data. It requires minimal or no-code experience and is extremely easy to create and use!

Vertex AI Search is accessible and offers a host of benefits for businesses of all sizes:

  • Find What You Need, Instantly: No more wasting hours digging through files. Vertex AI Search understands your queries and delivers precise results, even if your search terms aren’t an exact match.
  • Uncover Hidden Opportunities: Vertex AI Search doesn’t just find information; it reveals patterns, trends, and correlations you might have missed. This can lead to valuable insights for improving your products, marketing strategies, and customer service.
  • Make Data-Driven Decisions: With easy access to relevant information, your team can base decisions on solid evidence rather than guesswork.
  • Scale as You Grow: Vertex AI Search is a cloud-based solution, so it can easily adapt to your growing data needs.

Industry-Specific Use Cases: A Startup’s Playground

  • Ecommerce, Elevating the Shopping Experience: For e-commerce startups, personalized product recommendations are the key to unlocking customer loyalty and boosting sales. Vertex AI Search analyzes user behavior, purchase history, and preferences to deliver tailored suggestions. This boosts engagement, increases conversions, and fosters customer loyalty.
  • Healthcare, Understanding Medical Information: Vertex AI Search is transforming how patients and medical professionals access information. You can build AI-powered medical Q&A systems, and platforms that help healthcare professionals find relevant clinical trials. This not only empowers individuals but also aids in early disease detection and treatment.
  • Financial Services, Intelligent Financial Advisors: In the financial world, data is king. You can leverage Vertex AI Search to build intelligent search engines that assist customers with investment decisions, retirement planning, and financial goal setting.
  • Education, Personalized Learning at Scale: Vertex AI Search has the potential to revolutionize education. You can create AI-powered tutoring platforms that adapt to each student’s learning pace and style. Imagine a language learning app that uses Vertex AI Search to provide instant grammar corrections, vocabulary suggestions, and personalized practice exercises.
  • Content Creation and Marketing, Boosting Content Discovery: For startups in the content and marketing space, discoverability is crucial. Vertex AI Search can be used to build semantic search engines for blogs, news websites, or video platforms. This ensures that users find the most relevant content, even if their search terms don’t perfectly match the content itself.

Step-by-Step Guide

Google Cloud’s Vertex AI Search is designed to be user-friendly, even if you don’t have a team of data scientists. With its intuitive interface and pre-built connectors, you can get started quickly and easily. In this step-by-step guide, we will be building a search experience on publicly available sustainability reports obtained from Google Sustainability Reports.

Disclaimer: The data used in this demonstration is publicly available and does not represent any affiliation with or endorsement by any specific organization or entity.

Prerequisites

  • A Google Cloud Platform project with billing enabled.
  • Enable Vertex AI Agent Builder, BigQuery, Cloud Storage APIs.
  • [Optional] If you want to retrieve your data from Google Cloud Storage, make sure to upload your files to a Google Cloud Storage Bucket before creating the application.

Step 1:

On your Google Cloud Platform, Search for “Agent Builder” and select the “Agent Builder” option under Products & Pages.

Step 2:

Click on “+ Create App”.

Step 3:

Select “Search” as your app type and click “Continue”.

Note: Agent Builder also has other different types of applications depending on your need including:

  • Chat: For building conversational AI agents capable of answering complex questions and engaging in natural language interactions.
  • Recommendations: For creating content recommendation engines that can suggest relevant items to users based on their preferences and behavior.
  • Search: For building applications designed to retrieve information from a knowledge base or dataset.

This flexibility allows you to choose the app type that best aligns with your desired functionality within Agent Builder.

Step 4:

In the configuration panel, select the content type. For this application, I have selected “Generic”.

Note: Vertex AI Search offers different content options to tailor your search experience:

  • Generic: A versatile option suitable for a wide range of general-purpose search applications.
  • Media: Optimized for consumer-focused media platforms, including audio, video, and digital publishing.
  • Healthcare: Designed for handling clinical healthcare data in the FHIR (Fast Healthcare Interoperability Resources) format.

This flexibility allows you to choose the app type that best aligns with your desired functionality within Agent Builder.

Enable “Enterprise edition features” to activate extractive answers from documents, image search and website search. Also enable “Advanced LLM features” for search with follow ups and summarization!

Provide a suitable name for your application, and enter the name of your organization. Finally, select the region for your application and click on “Continue”.

Step 5: In the data section, select “+ Create Data Store”.

Note: Vertex AI Search offers diverse data source options, including Website URLs, BigQuery, Cloud Storage, Google Drive, and APIs. For the purposes of this guide, sustainability reports have been pre-uploaded to a dedicated Cloud Storage Bucket on Google Cloud Platform.

Choose “Cloud Storage” as your data source and specify the path to your bucket. You will need to indicate the data type you are importing (unstructured data, JSON, or CSV). Given that the bucket I have chosen contains PDF files, I have selected the “unstructured data” option. Proceed by clicking “Continue”.

Provide a name for your data store and click on “Create”.

Once the data store has been created, select the data store, and click on “Create”. Upon completion, all imported files will be accessible within the “Data” tab, under the “Documents” section.

Note: It may take several minutes to process the data depending on the number and size of the files.

Congratulations!!! You have now built your search engine application using Vertex AI Search on Agent Builder!!!

It’s Time to Test!

In the Preview tab, you can interact with your data by posing a wide range of questions, from summarizing document content to extracting specific numerical values.

For example, you could ask:

  • Summarization:Summarize the key findings of Alphabet’s 2022 environmental report.
  • Numerical values: “How many smelters were there in the year 2019?”
  • Specific details: “What initiatives did the company implement to reduce water usage?”

Integration of Different Components

You can integrate Vertex AI Search into your application using two flexible methods:

  1. API Integration: Embed search functionality directly into your website for a seamless user experience.
  2. Widget Integration: Utilize pre-built code snippets to customize your user interface or connect with your server-side application.

Choose the method that best aligns with your development workflow and requirements!

Key Takeaways

  • Agent Builder’s Vertex AI Search is a powerful tool that can be used to build personalized search experiences for various industries.
  • You can build your search engine application in just a few easy steps, and unlock hidden patterns and trends in your data — structured, unstructured or website data!
  • It can be used to build a variety of applications, including product recommendation engines, medical Q&A systems, and financial advisory systems which can be easily integrated with your own application.

Call to Action

👋 By the way, if you happen to be a startup owner who is actively seeking to propel your business to new heights with Cloud:

We invite you to join our exclusive virtual live workshops (links below), where you’ll gain hands-on guidance from Google Cloud experts and discover how to seamlessly integrate GCP into your operations. Don’t miss this limited-time opportunity to empower your startup with the knowledge and expertise needed to thrive in the cloud-driven world. Register now and secure your spot!

⚒️Startup Onboarding Workshop

⛑️Generative AI Workshop

🔑Data Analytics Workshop

🔐Security Workshop

📠Modern Applications Workshop

Disclaimer

This article is not a recommended practice for building and deploying production ready applications, only a fun way to explore the Google Cloud Platform and learn. Any opinions in this article are my own and not those of Google.

Google Inc makes no guarantee or warranty concerning the accuracy of said information and shall not be responsible for any loss or damage of whatever nature resulting from the use of, or reliance upon it. Google Inc does not guarantee that the use of any information contained herein will not infringe upon patent, trademark, copyright or rights of third parties.

It is your responsibility to follow the clean up instructions to stop all services to avoid incurring charges to your account.

--

--