Vertex AI Agent Builder를 사용하면 제한적인 머신러닝 기술을 가진 개발자라도 Google의 파운데이션 모델, 검색 전문 기술, 대화형 AI 기술을 활용하여 엔터프라이즈급 생성형 AI 애플리케이션을 만들 수 있습니다.
Vertex AI Agent Builder는 다음과 같은 기능으로 구성됩니다.
Vertex AI Agents
Vertex AI Search
Vertex AI Agents
Vertex AI Agents는 대규모 언어 모델(LLM)을 기반으로 빌드된 새로운 자연어 이해 플랫폼입니다.
대화식 사용자 인터페이스를 쉽게 설계하여 모바일 앱, 웹 애플리케이션, 기기, 봇, 대화형 음성 응답 시스템 등에 통합할 수 있습니다.
Vertex AI Agents를 통해 사용자가 제품과 상호작용할 수 있는 새롭고 흥미로운 방법을 제공할 수 있습니다.
Vertex AI Search는 대규모 언어 모델을 기반으로 하는 완전 관리형 플랫폼으로, 공개 또는 비공개 웹사이트 또는 모바일 애플리케이션을 위한 AI 지원 검색 및 추천 환경을 빌드할 수 있습니다.
앱 검색
Vertex AI Search를 사용하면 자체 데이터를 바탕으로 Google급의 검색 앱을 빠르게 빌드하고 웹페이지 또는 앱에 검색창을 삽입할 수 있습니다.
다음과 같은 다양한 유형의 검색 앱을 만들 수 있습니다.
일반 검색. 독점 데이터가 포함된 웹사이트 또는 데이터 스토어에 일반 검색을 적용하여 고객이 보길 원하는 콘텐츠에 Google 수준의 검색 환경을 제공하세요.
미디어 검색. 이 기능은 영화, 동영상, 음악과 같은 미디어 콘텐츠에 맞게 특별히 설계된 검색 기능입니다. 미디어 검색을 사용하면 잠재고객이 시청하거나 듣고 싶은 미디어 콘텐츠를 효율적으로 찾을 수 있습니다.
의료 검색. FHIR 데이터 스토어에 저장된 의료 기록을 쿼리할 수 있는 검색 기능입니다. Cloud Healthcare API FHIR 저장소에서 임상 데이터가 포함된 FHIR 리소스를 가져올 수 있습니다. FHIR 리소스에서 참조하는 이미지, PDF 파일, RTF 파일과 같은 구조화되지 않은 데이터도 검색할 수 있습니다.
추천 앱
사용자가 보고 있는 콘텐츠와 유사한 콘텐츠를 추천할 수 있는 최신 추천 앱을 자체 데이터로 빠르게 빌드할 수 있습니다.
다음과 같은 두 가지 유형의 추천 앱을 만들 수 있습니다.
미디어 추천. 동영상, 뉴스, 음악과 같은 미디어 콘텐츠를 추천합니다. 미디어 추천을 사용하면 최적화 목표를 사용하여 맞춤설정된 Google 품질의 결과를 통해 잠재고객이 보거나 읽을만한 콘텐츠 등 보다 맞춤설정된 콘텐츠를 찾을 수 있습니다.
일반 추천(미리보기). 미디어 외 콘텐츠에 대한 추천을 받습니다.
주요 특징
즉시 사용 가능한 자연어 이해 및 시맨틱 검색.
키워드 검색이나 패턴 일치를 실행하는 시스템을 구현하고 유지관리하지 않아도 고품질 검색 환경을 이용할 수 있습니다.
동의어 이해, 맞춤법 교정, 자동 추천 검색을 위한 즉시 사용 가능한 기능. 복잡한 자연어 처리 기술을 구현하지 않고도 사용자의 검색 환경을 개선할 수 있습니다.
생성형 AI. 구조화되지 않은 문서에 생성형 AI 기반 요약 및 대화형 검색을 적용하세요.
즉시 사용 가능한 추천. 최신 ML 기반 콘텐츠 및 메타데이터 이해를 통해 사용자가 보고 있는 콘텐츠와 유사한 콘텐츠를 빠르게 찾을 수 있습니다.
Vertex AI Agent Builder 콘솔 및 API 콘솔의 Agent Builder 페이지 또는 Google API를 사용하여 공개 웹사이트나 구조화된 데이터 또는 구조화되지 않은 데이터용 검색 앱 또는 추천 앱을 설정합니다.
즉시 사용 가능한 위젯. 웹사이트에 검색을 통합합니다. 자세한 내용은 웹페이지에 검색 위젯 추가를 참조하세요.
자기 학습. 자기 학습 순위 모델과 고급 분석을 활용하세요.
이를 위해서는 사용자의 클릭 스트림이 필요합니다.
미디어 최적화. 미디어 콘텐츠에 최적화된 추천 및 검색 앱을 만드세요.
의료 데이터에 대한 자연어 쿼리. 사전 쿼리 언어 지식 없이 FHIR 리소스를 검색합니다.
컨텍스트 인식 의료 검색. 구조화된 FHIR 검색에서 누락될 수 있는 시맨틱 관련성이 있는 검색 결과를 찾습니다.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2024-12-21(UTC)"],[[["\u003cp\u003eVertex AI Agent Builder empowers developers to create enterprise-grade generative AI applications using Google's foundation models, search capabilities, and conversational AI, even with limited ML expertise.\u003c/p\u003e\n"],["\u003cp\u003eVertex AI Agents, a natural language understanding platform, enables the design and integration of conversational user interfaces across various platforms, such as mobile apps, websites, and interactive voice response systems.\u003c/p\u003e\n"],["\u003cp\u003eVertex AI Search allows users to build AI-powered search and recommendation experiences for public or private websites and applications, including generic, media, and healthcare-specific search options.\u003c/p\u003e\n"],["\u003cp\u003eVertex AI Agent Builder provides features like out-of-the-box natural language understanding, semantic search, synonym recognition, spelling correction, auto-suggestions, generative AI capabilities for summarization and conversational search, and more.\u003c/p\u003e\n"],["\u003cp\u003eVertex AI Search also has the capability to build state-of-the-art recommendations on your data, with media and generic options, for content that users would appreciate.\u003c/p\u003e\n"]]],[],null,["# What is AI Applications?\n\nAI Applications lets developers, even those with limited machine learning\nskills, tap into the power of Google's foundation models, search expertise, and\nconversational AI technologies to create enterprise-grade generative AI\napplications for search, recommendations, and conversation agents.\n\nThe product AI Applications essentially packages the following aspects\ntogether:\n\n- Vertex AI Search: Search and recommendation apps, used for AI-enabled search, browse, grounded answer generation, and recommendations. See [Vertex AI Search](#es-and-personalize).\n- Dialogflow conversational agents: Conversational Agents, used for creating AI-driven conversational user interfaces, was originally available as part of AI Applications. In early 2025, these products have become distinct offerings. See the [Conversational Agents documentation](/dialogflow/cx/docs).\n\nThis user guide talks about Vertex AI Search and all its offerings in\ndetail.\n\n\n| **Product rebrand** : This product has been renamed since its introduction. Some of the former names include *Agent Builder* , *Vertex AI Search* (an existing subset of AI Applications), *Enterprise\n| Search* , and *Generative AI App Builder*. Regardless of the rebrand, the product's functionalities, such as search, recommendations, generative AI answers, and browse, remain true to the Google quality.\n\n\u003cbr /\u003e\n\nVertex AI Search\n----------------\n\nVertex AI Search is a fully-managed platform, powered by large\nlanguage models, that lets you build AI-enabled search and recommendation\nexperiences for your public or private websites or mobile applications. \n\n### Information retrieval using AI and LLMs\n\nVertex AI Search brings together the power of deep information\nretrieval, state-of-the-art natural language processing, and the latest in large\nlanguage model (LLM) processing to understand user intent and return the most\nrelevant results for the user.\n\nWith Vertex AI Search, you can build a Google-quality search app\non data you control. You also have the option to use the search results that you\nretrieve to ground generative AI LLM responses. For more information, see the\nblog post\n[Your RAG powered by Google Search](https://cloud.google.com/blog/products/ai-machine-learning/rags-powered-by-google-search-technology-part-1).\n\nWith recommendations, you can build a recommendations app across your data\nthat suggests content similar to the content that the user is viewing.\n\n### An easy experience to get started\n\nVertex AI Search makes it easy to get started with high-quality search\nor recommendations based on data that you provide. As part of the setup\nexperience, you can:\n\n- Use your existing Google Account or sign up for one.\n- Use your existing Google Cloud project or create one.\n- Create an app and attach a data store to it. Provide data to search or recommend by entering the URLs for your website content, importing your data from BigQuery or Cloud Storage, or importing FHIR R4 data from Cloud Healthcare API, or uploading through RESTful CRUD APIs. Syncing data from third-party data sources is available in Preview with allowlist. For more information, see [About apps and data stores](/generative-ai-app-builder/docs/create-datastore-ingest)\n- Embed JavaScript widgets and API samples to integrate search or recommendations into your website or applications.\n\n### Search apps\n\nWith Vertex AI Search, you can quickly build a Google-quality search\napp on your own data and embed a search bar in your web pages or app.\n\nYou can create the following different types of search apps:\n\n- **Custom search.** Apply custom search to websites\n or to data stores containing your\n proprietary data, giving your customers Google-quality search experiences on\n the content that you want them to see. For more information, see\n [Introduction to custom search](/generative-ai-app-builder/docs/about-generic-search)\n and [Get started with custom search](/generative-ai-app-builder/docs/try-enterprise-search).\n\n- **Media search.** This is a search capability\n specially designed for media content such as movies, videos, and music. With\n media search, audiences can efficiently find the media content that they\n want to view or listen to. For more information, see\n [Introduction to media search and recommendations](/generative-ai-app-builder/docs/about-media)\n and [Get started with media search](/generative-ai-app-builder/docs/try-media-search).\n\n- **Healthcare search.** This is a search capability that lets you query\n healthcare records stored in FHIR data stores. You can import FHIR resources\n that contain clinical data from your Cloud Healthcare API FHIR store. You can\n also search unstructured data, such as images, PDF files, and RTF files,\n referenced by the FHIR resources.\n\n### Recommendation apps\n\nYou can quickly build a state-of-the-art recommendations app on your own data\nthat can suggest content similar to the content that the user is viewing.\n\nYou can create the following two different types of recommendations\napps:\n\n- **Media recommendations.** Get recommendations for media content such as\n videos, news, and music. With media recommendations, audiences can discover\n more personalized content, like what to watch or read next, with\n Google-quality results customized using optimization objectives.\n For more information, see\n [Get started with media recommendations](/generative-ai-app-builder/docs/try-media-recommendations).\n\n- **Custom recommendations (Preview).** Get recommendations for non-media\n content. For more information, see\n [Get started with custom recommendations](/generative-ai-app-builder/docs/try-generic-recommendations).\n\n | **Note:** This feature is a Preview offering, subject to the \"Pre-GA Offerings Terms\" of the [GCP Service Specific Terms](https://cloud.google.com/terms/service-terms). Pre-GA products and features may have limited support, and changes to pre-GA products and features may not be compatible with other pre-GA versions. For more information, see the [launch stage descriptions](https://cloud.google.com/products#product-launch-stages). Further, by using this feature, you agree to the [Generative AI Preview terms and conditions](https://cloud.google.com/trustedtester/aitos) (\"Preview Terms\"). For this feature, you can process personal data as outlined in the [Cloud Data Processing Addendum](https://cloud.google.com/terms/data-processing-terms), subject to applicable restrictions and obligations in the Agreement (as defined in the Preview Terms).\n |\n | \u003cbr /\u003e\n |\n### Key features\n\n- **Out-of-the-box natural language understanding and semantic search.** Get a high-quality search experience without needing to implement and maintain systems that perform keyword searches or pattern matching.\n- **Out-of-the-box capabilities to understand synonyms, correct spellings, and\n auto-suggest searches.** Improve the user's search experience without the need to implement complex natural language processing techniques.\n- **Generative AI.** Get generative AI-powered summarization and conversational search for unstructured documents.\n- **Out-of-the-box recommendations.** Get state-of-the art, ML-based content and metadata understanding that lets your users quickly find content that is similar to the content that they're viewing.\n- **AI Applications console and APIs.** Use the [AI Applications](https://console.cloud.google.com/gen-app-builder/engines) page of the console or Google's APIs to set up a search app for your public websites or for your structured or unstructured data.\n- **Out-of-the-box widget.** Integrate search into your website. For more information, see [Add the search widget to a web page](/generative-ai-app-builder/docs/add-widget).\n- **Self learning.** Get self-learning ranking models and advanced analytics. This requires the user's clickstream.\n- **Optimization for media.** Create recommendation and search apps optimized for media content.\n- **Natural language querying of healthcare data.** Search FHIR resources without prior knowledge of any query language.\n- **Context-aware healthcare searches.** Find search results with semantic relevance that a structured FHIR search might miss.\n\n### Google Cloud console or the API?\n\nYou can implement Vertex AI Search in any of the following ways:\n\n- **Use the Google Cloud console.** Use the [AI Applications](https://console.cloud.google.com/gen-app-builder/engines) page of the console for a quick-start experience using a web interface. From the console, you can create your search app, import your data, test the user experience, and view analytics.\n- **Use the AI Applications API.** Use the AI Applications API when you're ready to integrate search or recommendations into your website or applications.\n- **Use both the Google Cloud console and the API.** You can set up your app and import your data using the console, for example, and then use the API to test the user experience and integrate it into your website or application."]]
