From Prompts to Autonomous Software Engineering Workflows in the Terminal
Build practical, real-world AI workflows directly from your terminal using the Gemini CLI. This course is designed for developers who want to move beyond simple prompt usage and start creating structured, repeatable, and autonomous coding workflows powered by modern AI systems.
What you’ll learn
- Build structured, agentic AI workflows using the Gemini CLI.
- Automate coding tasks like testing, debugging, and refactoring.
- Write effective task-oriented prompts for reliable AI output.
- Integrate AI into real development workflows using scripts and the terminal.
Course Content
- Introduction –> 6 lectures • 30min.
- Prompting for Agentic Workflows –> 3 lectures • 18min.
- Working with Files and Context –> 4 lectures • 21min.
- Testing, Debugging and Verification –> 5 lectures • 38min.
Requirements
Build practical, real-world AI workflows directly from your terminal using the Gemini CLI. This course is designed for developers who want to move beyond simple prompt usage and start creating structured, repeatable, and autonomous coding workflows powered by modern AI systems.
You will begin with a clear introduction to the Gemini CLI, including installation, authentication, and core interaction patterns. From there, the course quickly shifts focus to what actually matters: using the CLI as a foundation for agentic software engineering. Instead of treating AI as a one-off tool, you will learn how to design workflows where AI can plan, execute, iterate, and assist across multiple steps in your development process.
Throughout the course, you will build a series of progressively more capable workflows. These include structured prompt pipelines, task-oriented instructions, multi-step code generation, refactoring workflows, and automated scripting scenarios. You will see how to guide the model with precise instructions, manage context effectively, and reduce errors by breaking problems into smaller, composable steps.
A major focus is on integrating the Gemini CLI into real development environments. You will learn how to use it alongside shell scripts, existing tooling, and version control systems to automate repetitive tasks and accelerate development cycles. By the end of the course, you will be able to create workflows that can analyze codebases, generate features, transform data, and assist with debugging in a consistent and reliable way.
The course also covers best practices for working with agentic systems. This includes designing robust prompts, handling failure cases, validating outputs, and maintaining control over automated processes. You will understand when to rely on AI, when to constrain it, and how to structure workflows that remain predictable and maintainable over time.
This is a hands-on course. Each section includes practical examples and demonstrations that you can adapt to your own projects. Whether you are building internal tools, automating development tasks, or experimenting with AI-assisted engineering, the techniques taught here are directly applicable.
By the end of the course, you will not only understand how the Gemini CLI works, but how to use it as a core part of an agentic development workflow. You will be equipped to design systems that go beyond prompts and start delivering real, measurable productivity gains in your day-to-day engineering work.
If you are a developer looking to integrate AI into your workflow in a meaningful, structured way, this course will give you the tools and patterns to do it effectively.