Gemini for Google Cloud

Gemini for Google Cloud Cheat Sheet

Gemini for Google Cloud is a generative AI-powered collaboration product that provides assistance to a wide range of Google Cloud users, including developers, data scientists, and cloud architects. It is embedded in many Google Cloud products to offer an integrated assistance experience.

Product Offerings

Product	Type of Assistance	Key Tasks
Gemini Cloud Assist	Design, optimize, and operate cloud applications	Efficiently manage cloud applications; create, optimize, and troubleshoot resources quickly; manage resource security configurations
Gemini Code Assist	Develop, deploy, and troubleshoot code	Generate and debug code; generate unit tests; offer code optimization suggestions; build integrations with enterprise applications (Salesforce), databases (MySQL), and event-driven systems (Pub/Sub)
Gemini in BigQuery	Prepare and analyze data	Discover, transform, query, and visualize data with data canvas; explore data patterns with data insights; generate, complete, and explain SQL and Python code; get data preparation recommendations
Gemini in Colab Enterprise	Generate Python code in notebooks	Code completion suggestions while typing; code generation using the “Help me code” tool
Gemini in Looker (in Looker)	Analyze and export Looker data	Create custom visualizations; generate LookML modeling parameters; ask questions about data; build custom data agents; create calculated fields
Gemini in Looker (in Looker Studio)	Analyze and export Looker Studio data	Ask questions about data; build custom data agents; create calculated fields; export report visualizations to Google Slides
Gemini in Security Command Center	Simplify security investigations	Generate UDM search queries from natural language; summarize security cases; suggest resolution steps

Gemini Code Assist Editions

Feature	Standard Edition	Enterprise Edition
Intended audience	Customers with basic coding needs; organizations with strict data security requirements	Large enterprises with complex development processes; organizations wanting AI responses customized to private codebases
Code completion & generation	Available in supported IDEs (VS Code, JetBrains, Android Studio)	Same as Standard
Conversational chat in IDE	Available using opened files’ context	Same as Standard
Agentic chat	Prompt Gemini to complete complex, multi-step tasks using system tools and Model Context Protocol (MCP) servers	Same as Standard
Smart actions & commands	Right-click selected code; slash commands in quick pick	Same as Standard
Source citations	Available in IDE and Google Cloud console	Same as Standard
IP indemnification	Available	Same as Standard
VPC-SC & Private Google Access	Available	Same as Standard
Code customization	Not available	Available – get tailored suggestions based on your private codebases (GitHub, GitLab, Bitbucket)
Extended integrations	Firebase, Colab Enterprise, BigQuery data insights, Cloud Run, Database Studio	All Standard integrations PLUS Apigee, Application Integration, Gemini Cloud Assist

IDE Support

Gemini Code Assist works in the following IDEs:

VS Code
JetBrains IDEs (IntelliJ, PyCharm, and others)
Android Studio
Cloud Shell Editor
Cloud Workstations

How Gemini for Google Cloud Works

Training data: Gemini LLMs are trained on datasets of publicly available code, Google Cloud-specific material, other relevant technical information, and datasets used to train Gemini foundation models.

Contextual information: In the Google Cloud console, Gemini includes basic user context (project ID, product area, page title, organization ID) when giving responses. gcloud CLI commands and code may incorporate user context directly into generated snippets.

Data usage: Gemini does NOT use your prompts or responses for training or fine-tuning models. Data sent to Gemini is used strictly for serving responses and isn’t stored unless instructed by you.

Citations: When Gemini directly quotes at length from a web page or existing open source code, it cites that source. For code repositories, citations may reference applicable open source licenses.

Indemnification: Gemini for Google Cloud is covered as a Generative AI Indemnified Service. Google assumes certain responsibility for potential legal risks if you’re challenged on copyright grounds after using generated content.

Gemini Code Assist IDE-Specific Features

Feature	Description
Code completions	Receive suggestions as you type in the code editor
Code generation	Generate full functions or code blocks from comments
Unit test generation	Automatically create unit tests for your code
Debugging assistance	Get help understanding and fixing issues
Context sources	Gemini lists project files used as reference for responses (shown each time you use chat)
Citation prevention	You can prevent suggestions that match cited sources by adjusting IDE settings
Local codebase awareness	Uses large context window for in-depth understanding of your local project files
Agent mode	Complete complex multi-step tasks using system tools and MCP servers

Code Customization (Enterprise Only)

Get code suggestions based on your organization’s private codebase
Configure and use code customization through Gemini Code Assist settings

Setup

Product	Setup Guide
Gemini Cloud Assist	Set up Gemini Cloud Assist
Gemini Code Assist	Set up Gemini Code Assist
Gemini in BigQuery	Set up Gemini in BigQuery
Gemini in Colab Enterprise	Set up Gemini in Colab Enterprise