We provide course about Codex Beginners in english. 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.
Course description:
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
Course content:
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).
Instructors: Rogier Muller & Vasilis Tsolis
Rogier Muller is CTO of BlueMonks Group, an Amsterdam-based fintech compliance company, and co-founder of several companies. He`s a lifelong coder who moved early into AI-assisted software development. Today, he is the only person in the world to combine official ambassador roles across the three leading agentic engineering platforms: Cursor, Claude Code, and Codex.
Rogier has hosted numerous events worldwide and works closely with engineering teams, founders, and AI tooling companies on the practical adoption of agentic software development. His specific expertise is using agentic engineering in highly regulated environments, including but not limited to fintech, financial services, KYC/CDD, AML, GDPR, AFM-supervised contexts, auditability, data isolation, and compliance-heavy software delivery.
Vasilis Tsolis is a pioneer in document intelligence and agentic coding helping teams to change how they work across industries. He is an official Ambassador for Cursor, OpenAI Codex and n8n. He is the partner of Cognitiv+, an AI consultancy and software factory that helps organisations practically implement and adopt AI with enterprise confidence.
He has co-founded several companies and trains development teams across the US and EU on AI-assisted coding, with a consistent focus on integrating it into real workflows without losing control of the codebase. Background: engineering and law, twenty years across AI, construction, energy, and tech. Vasilis has worked with JPMorgan, Intel, PwC, and others along the way.
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
Language:
• The course is given in english
Course material:
The course fee includes course documentation, exercises, lunch and refreshments