Training course · OpenAI Codex · AI Engineering & LLM Applications

OpenAI Codex for AI Engineers

Delegate development work with OpenAI Codex: CLI and cloud agent, AGENTS.md, sandboxing, approvals and team PR workflows.

Duration
2 days (14 hrs)
Format
Online or on-site
Language
German or English
Participants
open dates or exclusive for your team
Certificate
Certificate of attendance

Overview

This course connects OpenAI Codex with the everyday processes of AI Engineering & LLM Applications. Participants work on realistic scenarios and leave with prompts, workflows and checklists they can use straight away.

OpenAI Codex is a coding agent that runs both locally, as a command-line tool and IDE extension, and as a cloud agent. Locally, Codex reads your repository, proposes changes and runs commands inside a sandboxed environment. In the cloud, it works on delegated tasks in parallel inside isolated containers and returns results as traceable diffs or pull requests. Over two days you will learn how to combine both modes effectively and how to scope tasks so that the agent can complete them reliably and with minimal back-and-forth.

Control and security sit at the heart of the course. With AGENTS.md you hand Codex project knowledge, build and test commands, and conventions. You configure approval modes and sandboxing so that the agent only does what you permit, and you set up cloud environments with suitable setup scripts. You will also integrate Codex into your pull request workflows, use automated code review, and define team rules on which tasks are delegated and how results are checked before they are merged into the main branch.

AI engineering is the discipline of turning language models into dependable products: chat assistants, document analysis, retrieval-augmented generation, agents with tool access and AI features inside existing applications. The code around the model is often more demanding than the model call itself, because prompt versioning, structured output, evaluation, cost control, latency, observability and protection against prompt injection decide success in production. AI coding assistants and agents accelerate exactly this work, from building evaluation pipelines and tool definitions to MCP servers and root-cause analysis in traces. In this course you learn to develop LLM applications systematically with AI support, make their quality measurable and run them securely, cost-effectively and in compliance with regulation.

The course is delivered live – either online in a virtual classroom or on-site in our training rooms in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart and Kaiserslautern. We can also run it exclusively for your team.

Who is this course for?

  • Software developers integrating LLM features into products
  • AI engineers and machine learning engineers
  • Data scientists focusing on generative AI
  • Backend and platform teams providing AI infrastructure
  • Technical product owners for AI features

Typical challenges we address

  • Prototypes shine in demos but deliver inconsistent quality in production
  • Without systematic evaluation, changes to prompts and models are risky
  • Costs, latency and rate limits spiral as usage grows
  • New security risks from prompt injection, data leakage and over-privileged agents

What you will be able to do

  • You use Codex locally and in the cloud for the development tasks each mode suits best.
  • You write an AGENTS.md that guides the agent reliably through your project.
  • You configure approval modes and sandboxing to match your security requirements.
  • You delegate tasks to the cloud agent and review the results systematically.
  • Build RAG applications, agents and LLM features with AI coding tools systematically and to production standard.
  • Make the quality of LLM applications measurable and manageable with automated evaluations.
  • Run LLM applications securely and cost-effectively, with observability, prompt injection defences and clear permissions.

Course outline

Day 1

Module 1 Codex at a glance: CLI, IDE and cloud
  • Comparing how Codex works locally and in the cloud
  • Installing the Codex CLI, signing in and starting it in a repository
  • Interactive sessions, non-interactive execution and model choice
  • Distinguishing tasks that suit a coding agent from those that do not
Module 2 Project knowledge with AGENTS.md
  • Building an AGENTS.md with build, test and lint commands
  • Stating conventions, architectural notes and prohibitions precisely
  • Using nested AGENTS.md files for monorepos and sub-projects
  • Checking effectiveness and refining instructions iteratively
Module 3 Sandboxing, approvals and security
  • Choosing approval modes deliberately, from read-only to largely autonomous
  • Understanding sandbox boundaries for file system and network access
  • Handling secrets, dependencies from the internet and prompt injection
  • Configuration files for team standards and safe defaults
Module 4 Delegating tasks to the cloud agent
  • Configuring cloud environments with setup scripts and dependencies
  • Writing tasks with a clear goal, context and acceptance criteria
  • Launching several tasks in parallel and comparing the results
  • Using logs and test output to understand the proposed solutions

Day 2

Module 5 PR workflows and code review in the team
  • Turning results into pull requests and requesting follow-up changes
  • Using automated Codex code review on GitHub
  • Embedding Codex in scripts and CI pipelines
  • Setting team rules for delegation, review duties and accountability
Module 6 Practice lab: AI Engineering & LLM Applications
  • Building RAG pipelines
  • Automating evaluation
  • Developing agents and tools
  • Securing structured output
  • Observability and cost
  • Defending against prompt injection
Module 7 Data protection, quality & transfer to daily work
  • LLM applications often process personal and confidential data: you learn to assess EU AI Act requirements on transparency and risk classes, design GDPR-compliant data flows with processing agreements and deletion policies, manage API keys securely and respect the licence terms of models and datasets.
  • Ground rules for safe AI use in your team
  • Checking results: facts, sources, tests and the four-eyes principle
  • Personal action plan for the next 30 days
  • Open questions, outlook and resources for further learning

Practical scenarios from your field

Building RAG pipelines

Implement ingestion, chunking, embeddings, vector search, hybrid retrieval and reranking, and generate answers with citations while honouring the access rights of source documents at retrieval time.

Automating evaluation

Create test sets with expected answers, define metrics for relevance, faithfulness and format, calibrate LLM-as-a-judge scoring and run evaluations in CI on every prompt or model change.

Developing agents and tools

Design tool definitions and MCP servers, implement agent loops with planning, tool calls and stop conditions, and build in permissions and human approval for consequential actions.

Securing structured output

Define JSON schemas for model responses, validate output, implement retry strategies on failure and make extraction tasks such as invoice or contract data robust. Faulty responses never reach downstream systems.

Observability and cost

Capture traces, token usage, latency and error rates, analyse problematic conversations and cut costs through caching, model routing and context reduction. Budgets and alerts per feature prevent surprises.

Defending against prompt injection

Test attack scenarios via manipulated documents, web pages or tool results, filter inputs and outputs, limit agent permissions and make red-teaming tests a fixed part of the pipeline.

Possible hands-on projects

Knowledge assistant with sources

A RAG chat over internal documents with citations, feedback buttons and an evaluation page that tracks answer quality across test questions.

Evaluation harness

A command-line tool that runs prompt variants against a test set, scores the results and clearly shows differences between versions.

Document extractor

A service that accepts PDFs, extracts structured data against a JSON schema, validates it and flags uncertain cases for manual review.

Prerequisites

You have hands-on programming experience, are comfortable with Git and the command line, and know pull-request-based workflows. You have solid programming skills, ideally in Python or TypeScript, and experience with APIs; prior exposure to LLM APIs helps but is not required.

Technical requirements

Your own laptop with a terminal, Git and a Node.js runtime; a ChatGPT plan with Codex access or an OpenAI API key, plus a GitHub account for the cloud exercises.

Data protection & compliance

LLM applications often process personal and confidential data: you learn to assess EU AI Act requirements on transparency and risk classes, design GDPR-compliant data flows with processing agreements and deletion policies, manage API keys securely and respect the licence terms of models and datasets.

Suggested dates

All dates are suggestions – your preferred date is always possible.

  • Tue, 24 November 2026 – Wed, 25 November 2026 Online Request
  • Mon, 30 November 2026 – Tue, 1 December 2026 Munich Request
  • Tue, 8 December 2026 – Wed, 9 December 2026 Online Request
  • Wed, 13 January 2027 – Thu, 14 January 2027 Hanover Request
  • Tue, 26 January 2027 – Wed, 27 January 2027 Online Request
  • Mon, 8 February 2027 – Tue, 9 February 2027 Berlin Request

Frequently asked questions

When should I use Codex locally and when in the cloud?

Locally you work interactively and close to the code, for example when debugging or making exploratory changes where you want to follow every step. The cloud agent suits well-defined tasks that can run independently, such as adding tests or fixing smaller bugs. You can start several of these in parallel and review each one as a pull request.

What belongs in a good AGENTS.md?

A good AGENTS.md explains how the project is built, tested and checked, which conventions apply and which areas must not be touched. It is concise, concrete and kept up to date. Notes on architecture and common pitfalls also help. During the course you write an AGENTS.md for a project of your own and test how it changes the agent’s behaviour.

How safe is it to let Codex run commands?

By default Codex works inside a sandbox with restricted file and network access and asks for approval before risky steps. You decide how much autonomy to grant through approval modes and configuration. In the course you learn to choose these settings deliberately and to limit risks such as prompt injection or running unfamiliar dependencies.

How does AI engineering differ from classic machine learning?

In classic machine learning you train your own models on your own data. In AI engineering you usually work with pre-trained language models via APIs or self-hosted models and focus on context, retrieval, tools, evaluation and operations. Statistics and training know-how matter less, while software engineering skills matter all the more. That is precisely where this course comes in.

How do we know whether our LLM application is good enough?

With an evaluation set of realistic questions and expected results, a mix of rule-based checks, LLM-as-a-judge and human spot checks, plus monitoring in production. The key is running evaluations automatically on every change. In the course you build such a setup and learn to calibrate judge models.

How much does “OpenAI Codex for AI Engineers” cost?

The course costs a flat €949 plus VAT per participant – whether online or on-site. This includes 2 training day(s) with a live trainer, course materials, a lab environment and a certificate of attendance.

Can I suggest my own date?

Yes. Our suggested dates are just a guide. In the request form you can always describe your preferred date or period in free text – we check availability and confirm within one working day.

Is the course online or on-site?

Both are possible: live online in a virtual classroom or on-site in Munich, Salzburg, Berlin, Augsburg, Hanover, Bremen, Stuttgart or Kaiserslautern. We can also run the course exclusively for your team.

More courses for AI Engineers

More OpenAI Codex courses

All 32 OpenAI Codex courses

OpenAI Codex · MunichOpenAI Codex · SalzburgOpenAI Codex · BerlinOpenAI Codex · AugsburgOpenAI Codex · HanoverOpenAI Codex · BremenOpenAI Codex · StuttgartOpenAI Codex · KaiserslauternOpenAI Codex · Online