Course · 5 chapters
Gemini CLI
Run Google's Gemini CLI as a real terminal coding agent: set up auth and Plan Mode, write GEMINI.md memory, drive subagents, and hand off to Jules. 5 chapters, practitioner-level, for engineers.
What you'll be able to do
- Set up Gemini CLI with Plan Mode
- Give the agent memory in GEMINI.md
- Run parallel subagents in worktrees
- Hand off long jobs to Jules
- Route Pro and Flash to cut token cost
- Pick the right billing for your team
What's inside
- 1Parcours Gemini CLI : Introduction
Ton orientation sur Gemini CLI de Google — ce que c'est, comment les chapitres s'articulent, et par ou commencer.
- 2Les essentiels de Gemini CLI
Maîtrise l'agent IA de codage open-source de Google pour le terminal — auth, modes d'approbation, Plan Mode et GEMINI.md.
- 3Gemini CLI Pro Flows
Sous-agents, extensions, worktrees et le handoff Jules — le parcours power-user d'avril 2026 pour Gemini CLI.
- 4Gemini CLI : maîtriser les coûts
Gouverne tes dépenses Gemini à travers CLI, Vertex, AI Studio, Code Assist et Jules — économie du contexte 1M, routage Pro vs Flash, choix de surfaces de facturation et gouvernance budgétaire.
- 5Maîtrise de la mémoire Gemini CLI
Conçois la persistance pour Gemini à travers le CLI, les sous-agents parallèles et Jules — hiérarchie GEMINI.md, stratégies de contexte 1M, Vertex AI Memory Bank et continuité des tâches asynchrones.
Frequently asked questions
- What will I learn in this Gemini CLI course?
- You learn to set up and run Google's Gemini CLI as a terminal coding agent: authentication, approval modes, Plan Mode, GEMINI.md memory files, subagents, extensions, git worktrees, the Jules async handoff, and cost control across billing surfaces. It runs 5 chapters totaling about 102 minutes.
- Who is this Gemini CLI path for?
- It is built for software engineers and developers who work in the terminal and want to use Gemini as a coding agent on real projects. The level is practitioner, so it suits people already comfortable writing code.
- Do I need coding experience to take it?
- Yes. This is a practitioner-level path for engineers, and the chapters assume you can read and write code and work in a command line. It does not teach programming from scratch.
- How long is the course, and is there a certificate?
- The path is about 102 minutes, roughly 1.7 hours, across 5 chapters covering intro, essentials, pro flows, cost control, and memory. On completion you receive an AI Academy certificate you can add to LinkedIn.
- How do I control Gemini CLI costs?
- A full chapter covers cost control: routing between Gemini Pro and Flash, the economics of the 1M-token context window, choosing among CLI, Vertex AI, AI Studio, Code Assist, and Jules billing surfaces, and setting budget governance.
Earn a certificate
Complete all chapters to receive your certificate of completion.