Codex Beginners

Codex has four surfaces (CLI, IDE extension, desktop app, Cloud) and most first-week users only use one of them. They also rarely understand sandbox modes or approval policies until something goes wrong. The default sandbox is workspace-write, which is the right answer for most work, but the failure mode is a developer flipping to danger-full-access the first time the agent gets stuck. This course replaces that pattern with a deliberate, surface-aware loop.

Course objectives

  • Pick the right Codex surface (CLI, IDE extension, app, Cloud) for a given task and explain why.
  • Read a repository well enough that the first prompt has the context the agent needs.
  • Write a spec with constraints and acceptance criteria before any code is generated.
  • Use AGENTS.md to make instructions durable across sessions.
  • Choose a sandbox mode and approval policy per task and defend the choice.
  • Decide whether to delegate, review carefully, or stay fully human-owned.

Target audience

Developers and senior developers joining a team that has standardised on Codex, or evaluating it seriously for the first time. Engineering managers who want to understand the loop their team is about to adopt are welcome but should expect to type.

Prerequisites

  • Has shipped code through a Git workflow before
  • Comfortable in a terminal
  • Has Codex CLI installed and authenticated with a ChatGPT account or API key before the course
  • Has completed the pre-course self-assessment

Module 1.
The Codex surface map. Quick walkthrough of CLI vs IDE extension vs desktop app vs Cloud. The mental model: CLI for terminal-native work, IDE extension when you want context inside VS Code or Cursor, desktop app for managing multiple agents on multiple projects, Cloud for long-running tasks that survive a closed laptop.

Module 2.
Reading the repo. Five minutes of repo reading saves forty minutes of agent thrashing. Lab: each participant writes a short context note for their own repo and uses it as the first message.

Module 3.
Spec-driven prompting. Constraints and acceptance criteria, not wishes. Lab: rewrite three weak prompts as specs, run both versions, compare diffs.

Module 4.
AGENTS.md, sandbox modes, approval policies. AGENTS.md is the durable instruction layer (also read by Cursor and other agents, so it is portable). Sandbox modes (read-only, workspace-write, danger-full-access) define what the agent can technically do. Approval policies define when it has to ask.

Lab: write an AGENTS.md and choose a default sandbox plus approval policy for the participant's own repo.

Module 5.
Verify, do not hope. Running tests, reading the diff, debugging with hypotheses instead of re-prompts. Lab: a seeded bug that passes the happy path.

Module 6.
Delegate, review, own. The decision model for what goes to Codex, what gets reviewed carefully, and what stays fully human-owned (security paths, migrations, architectural calls).

Practical information

Duration: 1 day
Price: 10 900 NOK
Language: English
Format: Classroom, virtual classroom, or in-company

FAQ

Hva er Codex Beginners?
Codex Beginners er et introduksjonskurs som lærer utviklere hvordan de jobber effektivt med Codex gjennom hele utviklingsløpet, fra å forstå kodebasen til å planlegge, generere og verifisere kode.

Hvem passer kurset for?
Kurset passer for utviklere og seniorutviklere som er nye i Codex, eller som ønsker en strukturert måte å bruke AI-assistert utvikling i praksis.

Trenger jeg forkunnskaper for å delta?
Ja, du bør ha erfaring med Git-basert utvikling, være komfortabel med terminal, og ha Codex CLI installert før kursstart.

Er kurset praktisk rettet?
Ja, kurset er hands-on og inkluderer flere øvelser hvor du jobber med egne eller realistiske kodebaser og oppgaver.

Hva lærer jeg som er nyttig i praksis?
Du lærer å velge riktig Codex-verktøy for oppgaven, skrive gode spesifikasjoner før kode genereres, jobbe sikkert med sandbox og godkjenningsregler, og verifisere og kvalitetssikre resultatet før det tas i bruk.

Other relevant courses

22. June
5 days
Classroom Virtual
23. September
3 days
Classroom Virtual
10. June
3 days
Classroom Virtual
18. August
4 days
Classroom Virtual