Cloud Hero: Gemini & Vertex AI

Overview

Gain practical experience using Vertex AI's Gemini models to tackle tasks such as code generation, function writing, and information extraction using natural language prompts. The course covers interacting with Gemini via cURL commands and the Vertex AI Python SDK, and culminates in deploying a Streamlit application that integrates Gemini Pro. It is designed to help learners modernize application development with generative AI.

Prerequisites

Participants should have some prior experience with Vertex AI.

Target audience

This course is designed for someone who has moderate experience with Vertex AI and is looking to modernize the way they develop applications using Vertex AI and Gemini.

Objectives

By the end of this course, learners will be able to:

• Install and use the Vertex AI Python SDK
• Interact with Gemini models using cURL and Python SDK
• Generate text, images, and video outputs from prompts
• Generate and use function calls from text prompts
• Extract information and entities from data using Gemini
• Integrate Gemini API with applications
• Build and deploy a Streamlit application on Google Cloud Run

Outline

Getting Started with the Vertex AI Gemini API with cURL

• Install the Python SDK
• Use the Gemini API in Agent Platform to interact with models
• Generate text from images, text prompts, and video using the model_name model

Introduction to Function Calling with Gemini

• Install the Google Gen AI SDK for Python
• Use the Gemini API to generate function calls from text prompts
• Help customers get information about products in the Google Store
• Call an external API to geocode addresses
• Extract entities from log data

Getting Started with the Vertex AI Gemini API and Python SDK

• Use the Google Gen AI SDK for Python to interact with generative AI services and models
• Connect to an API service
• Send text and multimodal prompts
• Set system instructions
• Configure model parameters and safety filters
• Manage model interactions, including multi-turn chat, content streaming, and asynchronous requests
• Use advanced features such as token counting, context caching, function calling, and text embeddings

Utilize the Streamlit Framework with Cloud Run and the Gemini API in Vertex AI

• Integrate Gemini API in Agent Platform with applications
• Build and deploy a sample application on Google Cloud Run
• Use the Streamlit framework to build a Cloud Run application

Exams and assessments

There are no exams or assessments associated with this course.

Hands-on learning

The course includes practical labs where learners install SDKs, interact with Gemini models, generate function calls, extract information, and build and deploy applications using Streamlit and Google Cloud Run.

Labs:

• Install the Python SDK
• Use the Gemini API in Agent Platform to interact with each model
• Use the model_name (model_name) model to generate text from image(s), text prompts and video
• Install the Google Gen AI SDK for Python
• Use the Gemini API in Agent Platform to interact with the model_name model:
• Generate function calls from a text prompt to help customers get information about products in the Google Store
• Generate function calls from a text prompt and call an external API to geocode addresses
• Generate function calls from a text prompt to extract entities from log data
• Install the Gen AI SDK.
• Connect to an API service.
• Send text and multimodal prompts.
• Set system instructions.
• Configure model parameters and safety filters.
• Manage model interactions (multi-turn chat, content streaming, asynchronous requests).
• Use advanced features (token counting, context caching, function calling, text embeddings).
• Integrate Gemini API in Agent Platform with applications
• Build and deploy the developed sample application on Google Cloud Run
• Use the Streamlit framework to build a Cloud Run application