We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Your developers go from first prompt to a reviewed, merged pull request in your own repository, and leave with a workflow that gets more of the backlog shipped.
Rebuilt on your team's work
Run privately for your team, on dates and at a length that suit you.
Every module ends with something participants can repeat on Monday morning in their own work and tools, so the team gets more done without adding people.
Take a ticket through to a reviewed pull request with Claude Code, without losing track of what changed.
Write project instruction files that make the agent follow your conventions instead of inventing its own.
Review agent-written code quickly, and recognise the failure modes that a green test suite does not catch.
Set permissions and hooks so the agent can work through a real repository without touching what it should not.
Turn recurring engineering chores into one-line commands and subagent runs.
Judge when reaching for an agent pays off, and when writing the code yourself is faster.
Most of the time goes on hands-on exercises. Module order and depth are set for your group before the first day.
Every length uses the same modules. A longer booking takes the team further through the same curriculum.
1 day
The agent loop and how to steer it. Enough for a team that should be writing real code with Claude Code by the end of the day.
2 days
Adds the working rhythm and the guardrails a team needs before it lets an agent near a shared repository.
4 days
Adds what only pays off once the whole team is on it: shared configuration, subagents, and measuring whether any of it worked.
We settle the length when we talk about your team, and it can still change afterwards.
Your version runs on your own repository: the CLAUDE.md is written for your codebase, the hooks enforce your review rules, and the ticket-to-pull-request exercise uses a real ticket from your backlog. If the code may not leave your cloud, we cover that setup first.
How a booking worksIn your office, with the team in the room. Best for a first rollout.
Live over video, in shorter blocks across several days.
A day on site to start, then shorter remote sessions on real work.
Every course runs privately for one team, on dates we agree together.
Every module can be rebuilt on your team's own work and tools. Tell me what the team works on and I will tell you what I would change.
AI agent trainer and software engineer
I am a software engineer before I am a trainer: I designed and built Tymeslot, an open-source scheduling platform in Elixir, and use Claude Code on production code like it. Claude Code was part of the programme I ran for engineering teams at Deutsche Telekom.
From engineering teams I trained to work with AI agents.
I worked closely with Luka as the topic sponsor for a workshop on integrating AI meaningfully into everyday development work.
He took the time to understand our participants' needs and adapted his approach to the tools and working environments available.
Participant feedback reinforced my own positive experience, with especially strong recognition of his subject expertise and the balance between theory and practice.
I wholeheartedly recommend Luka as a trainer to any team looking to build practical AI skills.
What teams ask before they book.
No. The course starts from how the agent loop works, so developers who have never run an agent can follow from the beginning. It does assume everyday professional coding experience, because most of the value sits in the review and judgement material.
Yes, and it is the version I recommend. Exercises land far better against code the team already argues about. Send me a rough description of the stack beforehand and I will adapt the exercises. Claude Code runs on your machines but sends your prompts and the code it reads to the model provider. If your code must stay within your own cloud or under stricter data terms, the course covers running Claude Code through Amazon Bedrock or Google Vertex AI, or with zero data retention, and I bring a comparable practice repository for anything that cannot be used at all.
Yes. Everyone needs working access on the day, so sort it a week ahead. Once we agree a date I send the exact requirements, including which plan covers what we will be doing. If procurement is slow at your end, say so early and I will plan the exercises around whatever access you can get in time.
On request, yes. A certificate naming the participant, the date and the material covered is available as an add-on and is sent digitally after the course. It records attendance; there is no exam.
Most of it, yes: the workflow, guardrail and review material carries straight over, because the hard part is context and review. The configuration and custom command sections are Claude Code specific. If your team has settled on Cursor or Codex, the Cursor and Codex courses cover the same ground for those tools.
Have a question about your team's setup?
Tell me what your team works on and how it uses AI today. A few lines are enough. I reply myself, usually within a day, and if this course is the wrong one I will point you to the one that fits.
In English or German, on site, remote or hybrid.
Claude Code Training
A few lines are enough
Engineering teams and providers I have delivered AI training for