APPLIED AI FOR SOFTWARE ENGINEERS
From Zero to Your First AI-Assisted Commit. The Foundations of OpenAI's Codex CLI.
A structured introduction to Codex CLI — OpenAI's terminal-native coding agent that reads your repo, plans changes, edits files, and runs commands locally while your source stays on your machine. Install and authenticate on macOS or Windows, run your first session, learn to read and approve diffs, use the essential slash commands, set up AGENTS.md project memory, and master the approval modes and sandbox that keep an AI agent safe.
Google Cloud Premier Partner
Install It. Run It. Ship Your First AI-Assisted Commit.
What You'll Build
- 💭
The Codex Mental Model
Understand the agent loop, OS-level sandbox, and human-in-the-loop approvals.
- 📦
Installed & Verified Codex
Install the scoped @openai/codex on macOS (npm/Homebrew) or Windows (PowerShell/WSL2).
- 🔑
Safe Authentication
Sign in with ChatGPT or use an OPENAI_API_KEY, handling keys securely.
- 💾
Your First AI-Assisted Commit
Run a session in a Git repo — intent, plan, approve, inspect diff, commit.
- ⌨️
Essential Slash Commands
/init, /model, /permissions, /review, /diff, /compact, /new, /quit.
- 📄
AGENTS.md Project Memory
Configure build, style, and boundary conventions Codex follows every session.
- 🛡️
Approval Modes & Sandbox
Read-only, auto, and full access — with safe rails as the everyday default.
- 🧪
Three-Part Practice Lab
A first script + README, an AGENTS.md setup, and the approval loop, hands-on.
Tools You'll Use
What You Will Build
Eight foundation modules and a three-part practice lab — from install to your first AI-assisted commit.
THE CODEX MENTAL MODEL
20 MINUnderstand what Codex CLI is — a terminal-native coding agent built on an agent loop, OS-level sandboxing, and human-in-the-loop approvals — and where it fits in your workflow.
INSTALLED & VERIFIED CODEX
30 MINCheck prerequisites (Node.js 18+, npm, Git), install the scoped @openai/codex (never the wrong unscoped package), on macOS (npm/Homebrew) or Windows (PowerShell/WSL2).
SAFE AUTHENTICATION
20 MINAuthenticate with ChatGPT sign-in (paid plan) or an OPENAI_API_KEY for automation/CI — treating the key like a password and rotating it on any leak.
YOUR FIRST AI-ASSISTED COMMIT
35 MINRun the core loop inside a Git repo: describe intent, review the plan, approve, inspect the diff, and commit or revert with confidence.
ESSENTIAL SLASH COMMANDS
25 MINThe everyday command set that drives a session: initialise, switch model, set permissions, review, diff, compact context, start new, and quit.
AGENTS.md PROJECT MEMORY
25 MINConfigure AGENTS.md so Codex follows your project's build, style, and boundary conventions automatically on every session.
APPROVAL MODES & SANDBOX
25 MINControl the three approval modes (read-only, auto, full access) and sandbox levels — defaulting to Auto + workspace-write and avoiding the --yolo bypass.
THREE-PART PRACTICE LAB
60 MINPut it together: build a first script + README, set up AGENTS.md, and practise the approval loop — then meet the completion criteria.
Your Instructors
Prem Kumar
AI Architecture Expert
"Anyone can build an AI demo. I teach you how to build the architecture behind systems that scale, because the gap between prototype and production is where most teams get stuck."
Specialty
Himal Rajan
Full Stack & AI Developer
"From React frontends to Python AI backends, I've spent my career building full-stack systems that actually ship, RAG pipelines, autonomous agents, and production APIs that handle real traffic. At RocketOne, I teach you the exact engineering decisions that turn an AI prototype into a system your users can depend on."
Specialty
Naveen Kumar
AI/ML Engineer
"I've built AI systems across Zoho, Virtusa, and now SquareShift, from machine learning pipelines to generative AI applications. At RocketOne, I teach you the practical skills to build AI-powered solutions that work in the real world, not just in notebooks."
Specialty
Industry practitioners with hands-on delivery experience at SquareShift Technologies, a Google Cloud Premier Partner serving Fortune 500 clients.
What You Need Before Day 1
Required
- Basic terminal familiarity (running commands, navigating folders)
- An OpenAI account (ChatGPT plan or API key)
- Node.js 18+, npm, and Git installed
- A laptop running macOS or Windows
Not Required
- Prior AI-tooling experience
- Advanced programming background
- Any paid Codex add-ons
This is the beginner (Level 1) track. No prior AI-tooling experience needed — only basic terminal familiarity and an OpenAI account.