Your engineers work faster with AI agents, and so do product, platform and security.
Seven hands-on agentic engineering trainings for teams, one or two days each, in the AI agent your team already uses. Mentoring is on call after every two-day training.
Book a call (opens in a new tab)- From EUR 750 per participant
- Remote or on site
Not sure where to start? Take the self-check
- Agent loop
- plan · build · verify
- Gates
- tests · security
- Delivery
- review · ship
- You steer
- intent · what ships
Production feedback shapes the next plan
- developers trained since 2025
- 1,000+
- teams trained since 2025
- 200+
Engineering and product teams we train across Germany, Austria and Switzerland, in German or English
From my point of view, day 2 was a success. I think it was very inspiring for all participants. The spark definitely caught, and that was the goal.
Florian Sauter (opens in a new tab) CTO EGYM
- Format
- Two days, in German or English, remote or on site in Germany, Austria and Switzerland. Agentic Engineering Security for Development Teams and Agentic Workflows for Business Teams take one day.
- Price
-
EUR 1,500 per participant
- One day:
- EUR 750 per participant
Up to 12 people per team, and you pay for 10 at most. Two days never cost more than EUR 15,000. On site, we add travel at cost. Net, plus VAT.
- Mentoring
- On call after every two-day team training: you book the hours your team needs, typically 2 to 4 a week over 4 to 12 weeks, at EUR 250 per hour, billed monthly as delivered. At those hours that is about EUR 2,000 to EUR 12,000 on top of the training. Optional with one-day trainings.
For engineering teams
-
Agentic Engineering for Developers and QA
Your whole team working with coding agents the same way.
- Software Engineers
- QA
- Analysts
- Who it's for
- Software and QA engineers, analysts and product owners.
- What your team takes away
-
- Project rules, agents and skills in your repository
- The handout, for everyone who attends
- Each participant gets their own certificate of attendance
- What the days cover
-
- Day 1: How AI coding agents work
- Day 2: Customize and connect
- Day 3 (optional): Roles
- Format
- Two days, remote or on site, handpicked trainer
-
Agentic Software Factory: Advanced Team Training
From each engineer's own setup to one pipeline the team shares.
- Software Engineers
- QA
- Who it's for
- Engineering teams that already use coding agents. You should know the basics from Agentic Engineering for Developers and QA, or already be fluent with agents.
- What your team takes away
-
- One delivery pipeline the whole team runs the same way, made of parts you can inspect: skills, subagents, state and contracts. First items from your backlog have run through it.
- A first factory, ready to grow toward production
- The handout
- Each participant gets their own certificate of attendance
- What the days cover
-
- Day 1: Your first software factory, small and observable
- Day 2: Review, gates and your own backlog
- Format
- Two days, remote or on site, handpicked trainer
-
Agentic Engineering for DevOps and Platform Teams
Routine platform work goes to agents, and your team approves every change.
- DevOps
- SRE
- Platform Engineers
- Who it's for
- DevOps, SRE and platform engineers who run infrastructure, pipelines and operations.
- What your team takes away
-
- Project rules and permission settings in your own infrastructure repository, and a first skill for a recurring chore
- A credentials setup that keeps write access to shared environments out of agent sessions
- A first change from your backlog as a pull request with its plan, ready for your normal review
- The training handout for every participant
- An individual certificate of attendance for every participant
- What the days cover
-
- Day 1: AI agents on your own infrastructure code
- Day 2: Your real work, through a safe change loop
- Format
- Two days, remote or on site, handpicked trainer
-
Agentic Engineering Security for Development Teams
Harden your agents, then use them to harden your code.
- AppSec
- Architects
- Software Engineers
- Who it's for
- Engineering teams already working with coding agents, with whoever owns security for them.
- What your team takes away
-
- A map of your own agent setups: every input, tool, credential and exit
- Your agent configuration tested against a planted prompt injection, with the first gaps closed
- A first draft of a threat model for one of your services, challenged by your team
- Where the hunt finds something: the finding with its reproduction and a first fix as a pull request for your normal review
- Security rules for your agents' instruction file and the security gate your team chose as its next step
- An individual certificate of attendance for every participant
- What the day covers
-
- Day 1: Securing AI agents and hunting with them
- Format
- One day, remote or on site, handpicked trainer
For product and leadership
-
Agentic Engineering for Product Managers and Product Owners
Specs your engineers' agents can build from.
- Product Managers
- Product Owners
- UX
- Who it's for
- Product managers, product owners and UX. For every background: everyone works hands-on with an agent.
- What your team takes away
-
- Specs your engineers' agents can act on
- One vocabulary shared with your engineering team
- First product workflows in your team's coding agent and a starter kit of prompt templates and agent configurations
- An individual certificate of attendance for every participant
- What the days cover
-
- Day 1: From idea to prototype, then into the coding agent
- Day 2: Specs, priorities and roadmaps with AI agents
- Format
- Two days, remote or on site, handpicked trainer
-
Agentic Engineering for Engineering and Product Leaders
What to steer, what to measure, and what to leave alone.
- CTOs
- VPs
- Engineering Managers
- Who it's for
- One company's leadership group: CTOs, VPs, heads of engineering and product, and engineering managers.
- What your team takes away
-
- A shared vocabulary with your engineering and product teams
- A view of what to measure and what to leave alone
- An individual certificate of attendance for every participant
- What the program covers
-
- Day 1: Where you stand, what AI agents change
- Day 2: Leading the change
- Format
- Typically two days, remote or on site
For business teams
-
Agentic Workflows for Business Teams
Teams outside engineering hand recurring work to agents, and a person still makes the call.
- Marketing
- Finance
- Sales
- HR
- Service
- Who it's for
- For everyone outside engineering and product: marketing, finance, sales, HR, legal, customer service. Any experience level.
- What your team takes away
-
- Every participant's recurring task written down, plus a first skill for it
- Team rules for agents: what they can see, which account they use, where a person approves
- A shortlist of your teams' tasks that suit an agent, each with a named owner
- An attendance certificate per participant, to document the training as an AI literacy measure under Article 4 of the EU AI Act, which names no specific measure
- What the day covers
-
- Day 1: From one recurring task to a working AI agent
- Format
- One day, remote or on site, handpicked trainer
Training days that convince even sceptical groups
-
Training
I read the summary of the training and I'm really happy with the content. On top of that, the training and you as the trainer went down very well with the participants. That says a lot, because in my view this was our toughest group.
Alexander Heusingfeld (opens in a new tab)
Head of Platforms & Lifecycle Management Vorwerk -
Training
Thanks again for the great workshop. We enjoyed it very much and took a lot out of the in-depth discussion. The links and materials are very helpful.
Daniel Gebler (opens in a new tab)
CTO Picnic
-
Clients
Tegtmeier Internet Solutions
Two training days on their own code in January 2026, then weekly mentoring from April to June. Three months after the training, the Managing Director put it simply: everyone here develops agentically.
-
Clients
A large DIY retailer in Germany
Consulting since September 2025, team trainings across the retailer's product areas since May 2026, and a bootcamp on a context engine for several million lines of legacy code.
Everyone has an AI agent working before lunch on day one
-
01
Prep call
Before the training, we agree with you which topics go deep and which stay light. We start from what your team already has, and tailor the training to it.
-
02
Training days
Before lunch on day one, everyone has an agent at work, in an example repository or your team's own code.
-
03
Practice between the days
Back to back or a week apart, with practice on your own backlog in between.
-
04
Mentoring
Your team applies the training to its own tasks. A mentor joins refinement, planning, implementation and review where it helps.
What every agentic engineering training has in common 8
What every agentic engineering training has in common
-
Understand the mechanics first
Our mental models show plainly how an agent works. That understanding is what lets people use agents to the full.
-
Everyone comes along
Every group has people far ahead and people far behind. We bring them all along: a team needs a shared language and a shared base more than one fast member.
-
We teach a field still being worked out
Our trainers say where they disagree and why. In at least one live session, they take on a real task from your team as it comes up.
-
Your agent is the example, new ways of working the subject
Claude Code, Codex, GitHub Copilot, Gemini CLI or Cursor: the names will change. The way of working stays, whatever ships next year.
-
Type less, delegate more
The role changes. People learn to hand work over so the agent checks its own work first.
-
Context engineering over prompt engineering
With the right context, agents make good decisions. Teams learn to provide it.
-
Learning from failure
Everyone learns to spot a failed agent run and put it to use, and that failed experiments are worth sharing.
-
You keep the material
After the training, your company keeps the slides, exercises and handouts.
Engineering, product and leadership, all working with agents the same way
Engineering, product and leadership, all working with agents the same way
| Before | After |
|---|---|
| Engineers | |
| Before: Engineers write every line themselves. | After: Engineers hand the task to the agent and review the result. |
| Before: Engineers state what they want, then check the result themselves. | After: Before it starts, engineers settle with the agent how it checks its own solution. |
| Before: The agent runs with full permissions. | After: The team starts with narrow permissions. Secrets and production-adjacent files stay explicitly blocked. |
| Before: Each session starts with the project rules, explained again. | After: CLAUDE.md holds project rules and safe test commands. The team keeps it current. |
| Before: The agent is handed a whole module to modernize in one go. | After: The agent takes small steps. An engineer reviews each diff before the commit. |
| Product | |
| Before: Engineers rewrite product's tickets before an agent can use them. | After: Product writes specs the engineers' agents can work from. |
| Before: Product and engineering name the same work differently. | After: Product and engineering use the same words for the same work. |
| Leaders | |
| Before: Leaders hear about agents in their teams' reports. | After: Leaders speak the same language as engineering and product. |
| Before: Any new metric about AI use gets tracked. | After: Leaders know what is worth measuring and what is not. |
What the organization gains: shorter lead time from idea to first users, and lower cost per change shipped.
What teams ask before they book
We already have Copilot or Cursor licences. Why train?
Licences put the tool on every machine. Often developers then use it a little, report mixed results, and delivery stays the same. The training changes how the team works: how an agent works, what context it needs, and how to delegate so it checks its own work. That carries over to the licences you have: the training runs directly on them, in Copilot, Cursor or any other major coding agent.
Which tools do you teach?
Our engineering, product and leadership trainings run in the coding agent your team already uses: Claude Code, Codex, GitHub Copilot, Gemini CLI or Cursor. We have delivered each of these trainings in every one of these agents. Agentic Workflows for Business Teams uses a desktop app: Claude, ChatGPT or Microsoft Copilot.
The instruction files, settings and approval modes differ by agent. We set them up with your team on day one. The practice is Agentic Engineering, and it carries over to whatever your company standardises on next.
No coding agent yet? We bring one that fits your team. Licences are on you, whether you supply them or we do.
Will this work on our legacy code?
Yes. Most teams start in a prepared example repository, then move to their own code, legacy included: the code they will keep maintaining afterwards.
How many people per group?
Eight to twelve people per group. The exercises are built for that size, and smaller or larger groups work less well in our experience. Participants eleven and twelve cost nothing extra. Fewer than eight? You pay the same EUR 750 per participant per day, or bring a neighbouring team along. A larger team splits into several groups.
EnablementTraining a large engineering organizationRemote or in person?
Both. Remote runs on your Microsoft Teams or Google Meet, cameras on, in a shared repository. On site, we come to you and bill travel and accommodation at cost.
Is funding available?
German companies may qualify for support under the Qualifizierungschancengesetz, depending on the company and the participants. We look at the route with you.
What happens after the training?
Mentoring is on call after every two-day team training: you book the hours your team needs, typically 2 to 4 a week over 4 to 12 weeks, at EUR 250 per hour, billed monthly as delivered. At those hours that is about EUR 2,000 to EUR 12,000 on top of the training. We stay with your team while it puts what it learned to the test in real work.
Teams that want agents to take work end to end can continue with a Factory Pilot and rollout.
Procurement and legal?
Your procurement and legal teams find everything they need on one page.
AboutProcurement and legalStart with one team
Tell us who should join and where your team is today. Within one working day, you get an offer, or our advice on which training to start with.
Elroy drafts the email with you in Claude or ChatGPT, and only you can send it.
Book a call (opens in a new tab)