Gemini CLI: Code & Create with an Open-Source Agent
by DeepLearning.AI
A short video course on using Google's open-source Gemini CLI agent for coding, data analysis and content work from the terminal.
Overview
This DeepLearning.AI short course, built with Google and released on January 21, 2026, teaches Gemini CLI, an open-source agentic coding assistant that runs in the terminal with access to your local filesystem, development tools and cloud services. Jack Wotherspoon, a Senior Developer Advocate at Google who works on Gemini CLI, teaches it, and Andrew Ng introduces it. The course runs 1 hour 23 minutes across 11 video lessons: an introduction, what Gemini CLI is, a tour of the tool, context with GEMINI.md, workflows with the Model Context Protocol, customisation with Gemini CLI extensions, then four applied lessons on software development, data analysis, content creation and learning, and a conclusion. The running project is an AI conference: learners build a website session catalog, a data dashboard that combines local CSV files with Google Sheets, and social media and blog drafts generated from recordings. Along the way the course covers automated code review with GitHub Actions and multi-tool workflows that reach Google Workspace and Canva. DeepLearning.AI lists it as Beginner and recommends basic command-line and Python familiarity. Videos are free; the graded assignment and certificate need a PRO membership. On the course forum, learners reported thin installation guidance, and staff pointed them to the course GitHub repo and the official Gemini CLI docs.
At a Glance
- Topic
- Agentic
- Level
- Beginner
- Format
- Course
- Cost
- Freemium
- Duration
- 1 hour 23 minutes of video, 11 lessons plus 1 graded assignment, self-paced
- Provider
- DeepLearning.AI
- Hands-on
- Yes — code/exercises
- Certificate
- Available
What You’ll Learn
- ✓Install Gemini CLI and use it as an agentic coding assistant in the terminal
- ✓Give the agent persistent project context and memory with a GEMINI.md file
- ✓Connect Model Context Protocol servers to build multi-tool workflows across local files and cloud services
- ✓Customise Gemini CLI with extensions, including Google Workspace integrations
- ✓Build web front-end features and automate code reviews with GitHub Actions
- ✓Analyse local CSV files and Google Sheets data to build a combined dashboard
- ✓Generate blog posts and social media drafts from audio or video recordings
Highlights
- •Taught by a Google Developer Advocate who works on Gemini CLI itself
- •Short enough to finish in an evening: 1 hour 23 minutes across 11 lessons
- •Covers GEMINI.md context files, MCP servers and extensions, which carry over to other terminal agents that use similar patterns
- •Applies the agent to data analysis and content creation as well as coding, all within one conference project
- •Learner feedback on the course forum flags weak setup instructions, so plan to use the course GitHub repo and official docs alongside the videos
Who It’s For
Best For
- ✓Developers trying a terminal coding agent for the first time
- ✓Engineers evaluating Gemini CLI against other coding agents
- ✓Teams wanting to automate code review and routine development tasks
- ✓Knowledge workers who want an agent for data analysis and content drafting
Prerequisites
- •Basic command-line experience
- •Basic Python familiarity is recommended but not required
FAQ
What is Gemini CLI: Code & Create with an Open-Source Agent?
A DeepLearning.AI short course, taught by Google Senior Developer Advocate Jack Wotherspoon, on Gemini CLI, Google's open-source terminal coding agent. It suits developers new to agentic coding tools who want to set up GEMINI.md context, connect MCP servers and extensions, and run multi-step development and data tasks from the command line.
Is Gemini CLI: Code & Create with an Open-Source Agent free?
Gemini CLI: Code & Create with an Open-Source Agent offers free content, with paid options for certificates or premium features.
What level is Gemini CLI: Code & Create with an Open-Source Agent for?
Gemini CLI: Code & Create with an Open-Source Agent is aimed at a beginner audience. Recommended background: Basic command-line experience, Basic Python familiarity is recommended but not required.
How long does Gemini CLI: Code & Create with an Open-Source Agent take?
Expect roughly 1 hour 23 minutes of video, 11 lessons plus 1 graded assignment, self-paced. Most learners work through it at their own pace.
What will I learn from Gemini CLI: Code & Create with an Open-Source Agent?
You'll learn: Install Gemini CLI and use it as an agentic coding assistant in the terminal; Give the agent persistent project context and memory with a GEMINI.md file; Connect Model Context Protocol servers to build multi-tool workflows across local files and cloud services; Customise Gemini CLI with extensions, including Google Workspace integrations; Build web front-end features and automate code reviews with GitHub Actions; Analyse local CSV files and Google Sheets data to build a combined dashboard; Generate blog posts and social media drafts from audio or video recordings.
Topics
Sources
This page was written from 4 sources, 3 on domains other than deeplearning.ai.