Agentic AI Enablement: the Deep Dive.
As in the Agentic AI Enablement course you get a detailed introduction to the world of agentic coding environments. In the full-day workshop you experience in practical, self-conducted exercises how you realise your own sample project. Guided by the coach you learn the fundamentals of Claude Code and ChatGPT Codex and see at every step how a coding agent plans, builds and tests.
What separates the practice day from the introduction.
Build it yourself from the first minute
Installation, sign-in and the GitHub connection happen on your own machine. After the first module the agent answers in your environment; every further exercise builds on that.
A project that belongs to you
The exercise project Mini-Markt takes shape in your own repository. Briefing, tests, two features and the GitHub finale are all driven by you, in your own words instead of prescribed click paths.
Close coaching in a small group
A maximum of twelve participants, so no blocker is left behind. Whoever gets stuck shares the screen; the coach untangles the knot within the same exercise instead of postponing it to the break.
Moves for the everyday work after
Slash commands, subagents, hooks, a custom command for the team and the design handover. By the end you have executed each of these workflows at least once yourself, not just seen them.
How the day unfolds.
The map and your own environment
A compact map of both platforms, then the building starts: desktop app, CLI and GitHub connection on your machine.
- Models and subscriptions of both vendors in fast forward
- Installation and first start in the exercise folder
- Walk through the approval modes on your own setup
- GitHub connection with your own permissions
CLAUDE.md: the project memory
The agent scaffolds your project itself, then it gets a memory that passes the read-aloud test.
- Project scaffold as the first real agent task
- Read and sharpen the /init draft
- Phrase verifiable rules, with the strike-through test
- Remember-on-request for everyday work
Feature development on your own project
The centrepiece: Mini-Markt from a briefing, tests with a genuine red moment and two features in the rhythm of goal, plan, diff, test.
- Send the briefing, read the plan, approve deliberately
- Understand a red test first, then have it fixed
- Phrase feature B in your own words and drive it to green
- Finale: repository on GitHub, /code-review, pull request
Toolbox, custom command and design
The most important slash commands in real sessions, subagents and hooks, then your first custom command takes shape. To close, the design handover.
- Commands from /usage to /loop, each with its Codex counterpart
- The hook principle: enforced instead of requested
- Create a custom command in .claude/commands
- Hand a design draft over to Claude Code
What you can do after the workshop.
What it costs.
What is included
- Full-day hands-on workshop with close coaching
- Your own exercise project from setup to the merged pull request
- Slides, cheat sheet and reference repository
- Confirmation of attendance
Discounts
Early-bird discount 10% when booking more than 30 days before the date. Group discount 10% from 5 participants registered together.
In-house training?
For an on-site learning environment and a format that you can run repeatedly for your colleagues, we deliver this workshop in-house as well. Reach out for a tailored offer.
Frequently asked.
Why do I need an active Pro subscription?
You work in your own tool for the whole day. Claude Code does not start without a Pro or Max subscription, and the Codex free limits do not carry through a full exercise day. One month of Claude Pro or ChatGPT Plus is enough and can be cancelled monthly.
Is my laptop enough?
A normal developer laptop with admin rights for installations is sufficient. You need PowerShell or a terminal, a browser and a GitHub account. The account can be created in two minutes beforehand.
What separates the Deep Dive from the introduction?
The distribution of roles. In the four-hour introduction the coach conducts all exercises himself; a screen to watch is all it takes there. In the Deep Dive you are at the wheel: every exercise runs on your machine, the coach accompanies you and steps in on blockers. That is why the Pro subscription is a prerequisite, and why you take home your own repository along with practised workflows.
I work with Codex instead of Claude. Does the workshop fit?
Yes. The exercises run in Claude Code, and every concept comes with its Codex counterpart. If you stay on your Codex subscription you transfer the patterns directly; the vocabulary mapping is part of the workshop.
Is attendance documented?
Yes, all participants receive a confirmation of attendance listing the covered content. It works for internal records and HR files.
A day at whose end your agent is running.
A working environment, a project memory with verifiable rules, two features built by you and a repository you keep.
Maybe a different pillar fits your situation better.
Quality Consulting
Strategie, Methodik, Frameworks für belastbare Qualität. Audits, Konzepte, AI-Compliance.
→Quality Services
Operative Test-Manpower, Interim-Testmanagement und Vermittlung aus dem Fachnetzwerk.
→Quality Education
Workshops, Schulungen und 1:1-Coaching für Test-, Projekt- und KI-Compliance-Themen.
→CT Map
Übersicht aller drei QCT-Säulen mit Wegweiser zu deinem passenden Einstiegspunkt.
→