Tool or Technology
TT0155

Codex in Action

Build, Debug & Ship Software with AI Agents

Codex
Duration
16 Hours
Level
Intermediate
Delivery
Instructor-Led
Format
Capability Training

Programme Objectives

  • Develop functional proficiency in using Codex to understand repositories, generate code, implement features, refactor applications, debug issues, run tests, and review software changes.
  • Apply effective prompting, context engineering, repository instructions, and planning techniques to improve Codex performance on realistic engineering tasks.
  • Work effectively across Codex CLI, IDE, ChatGPT, cloud, and GitHub-based development workflows.
  • Use Skills, AGENTS.md, MCP, and multi-agent workflows to create reusable, project-aware AI-assisted engineering processes.
  • Apply appropriate testing, code review, sandboxing, permissions, security, and human-approval practices when delegating software-development work to AI agents.

Technology covered

OpenAI CodexCodex in ChatGPTCodex CLICodex IDE ExtensionCodex CloudGitHub IntegrationCode ReviewAGENTS.mdCodex SkillsMCPWeb SearchSandboxed ExecutionApproval ModesMulti-Agent Workflows

Who should attend

  • Software Developers
  • Software Engineers
  • Frontend Developers
  • Backend Developers
  • Full-Stack Developers
  • DevOps & Platform Engineers
  • QA & Test Automation Engineers
  • Application Support Engineers
  • Technical Leads
  • Solution Architects
  • Engineering Managers
  • AI-Assisted Development Teams

Prerequisites & Participant Readiness

  • Working knowledge of at least one programming language
  • Familiarity with functions, classes, APIs, application architecture, and source-code structures
  • Basic experience with Git and repository-based development
  • Familiarity with command-line or terminal usage is recommended
  • Basic understanding of testing and debugging practices
  • Experience using an IDE such as VS Code or a compatible environment is beneficial
  • Basic understanding of Generative AI and prompting is helpful
  • Access to Codex and an approved development repository should be available for practical exercises
  • Completion of shorter Codex programmes is not required

TOC Modules

Concepts
  • Understanding Codex as an AI software-engineering agent
  • Understanding the difference between AI code completion, conversational coding, and agentic development
  • Understanding Codex across ChatGPT, CLI, IDE, cloud, and repository workflows
  • Understanding how Codex reads files, modifies code, executes commands, and verifies work
  • Understanding developer responsibility when delegating implementation tasks
Practical activities
  • Exploring the Codex development environment
  • Connecting Codex to a sample development project
  • Asking Codex to explain the structure and purpose of an unfamiliar repository
  • Comparing manual and agent-assisted approaches to a small engineering task

Scenarios

Business Requirement to Production-Ready Feature

Requirement → Repository Analysis → AGENTS.md Context → Implementation Plan → Codex Development → Tests → Code Review → Pull Request → Developer Approval

Participants use Codex throughout a realistic feature-development lifecycle, from understanding an unfamiliar repository and planning changes through implementation, verification, review, and final human approval.

Production Defect to Verified Resolution

Production Issue → Logs & Error Context → Codebase Investigation → Root-Cause Analysis → Agent-Assisted Fix → Regression Testing → Code Review → Verified Resolution

Participants use Codex to diagnose a real-world software defect, trace the problem across relevant components, implement a controlled fix, verify the solution, and prepare the change for safe integration.

Related programmes

Continue with programmes from the same capability area.

Take the next step

Ready to make this programme work for your team?

Customise modules, duration and business scenarios for your team.

Instructor-ledVirtualHybrid

Designed around your roles, tools and real workflows.